CN116384157A

CN116384157A - Land utilization change simulation method

Info

Publication number: CN116384157A
Application number: CN202310603066.9A
Authority: CN
Inventors: 高培超; 张潇丹; 高怡凡; 宋长青
Original assignee: Beijing Normal University
Current assignee: Beijing Normal University
Priority date: 2023-05-26
Filing date: 2023-05-26
Publication date: 2023-07-04
Anticipated expiration: 2043-05-26
Also published as: CN116384157B

Abstract

The invention provides a land utilization change simulation method, which relates to the technical field of data processing and comprises the following steps: acquiring the land system type of each subarea in the target area, each variable driving factor corresponding to the target area and each land service demand; determining suitability of each land according to each land system type and each change driving factor, and calculating service capacity of each land; determining each competitive advantage according to each land system type, each land service capability and each land service demand; according to the land utilization current situation data of the target area, calculating each conversion resistance and each neighborhood influence; calculating the conversion potential of each land according to the suitability of each land, each conversion resistance, each neighborhood influence and each competitive advantage; the improved land utilization model is input into the land utilization current system data, the land conversion potential, the land service capacity and the land service demand to obtain a simulation result, so that the supply and demand balance can be realized, and the efficiency and the reliability of land utilization change simulation are improved.

Description

Land utilization change simulation method

Technical Field

The invention relates to the technical field of data processing, in particular to a land utilization change simulation method.

Background

Land utilization change simulation is an important component in land change science, and can infer future land space patterns according to the current or specific time development rules. By analyzing different future development scenario assumptions, the influence of different development scenario assumptions, plans, policies and the like on the land space pattern can be judged by using land utilization change simulation.

In the prior art, a clirondo model is typically used for land use variation simulation. However, the Logit regression method is used by the CLUMondo model to establish the linear relation between the suitability parameter and the biophysical and socioeconomic factors, which is not fit to the actual situation and affects the accuracy of the simulation result. Accordingly, there is a need for an effective method to solve the above-mentioned problems.

Disclosure of Invention

Aiming at the problems existing in the prior art, the embodiment of the invention provides a land utilization change simulation method.

The invention provides a land utilization change simulation method, which comprises the following steps:

acquiring the land system type of each subarea in a target area, each change driving factor corresponding to the target area and each land service demand;

determining the land suitability of each subarea according to each land system type and each change driving factor, and calculating the land service capacity of each land system type;

Determining the competitive advantage of each subarea according to each land system type, each land service capability and each land service demand;

calculating conversion resistance and neighborhood influence of each land system type according to the land utilization current situation data of the target area; calculating the land transformation potential of each subarea according to the suitability of each land, the transformation resistance, the influence of each neighborhood and the competitive advantage;

and inputting the land utilization current data, the land conversion potential, the land service capacity and the land service demand into an improved land utilization model to simulate land utilization change, so as to obtain a simulation result.

According to the land utilization change simulation method provided by the invention, the land system type of each subarea in the target area is obtained, and the land utilization change simulation method comprises the following steps:

determining the land utilization type of each subarea in the target area according to the land utilization current situation data;

determining the area occupation ratio of each subarea corresponding to the land utilization type aiming at each land utilization type, wherein the area occupation ratio represents the ratio of a target area to the total area of the subareas, and the target area represents the land area belonging to the land utilization type in the subarea;

And determining the land system type of each subarea in the target area according to each land utilization type and each area occupation ratio.

According to the land use change simulation method provided by the invention, the land system type of each subarea in the target area is determined according to each land use type and each area occupation ratio, and the land system type comprises the following steps:

for each land utilization type, calling a natural breakpoint algorithm for the area ratio of each subarea corresponding to the land utilization type, and determining a first threshold value and a second threshold value;

dividing the land utilization type into three land system types according to the first threshold value and the second threshold value;

and comparing the area proportion of each subarea corresponding to the land utilization type with the three land system types respectively, and determining the land system type of each subarea corresponding to the land utilization type.

According to the land utilization change simulation method provided by the invention, the land suitability of each subarea is determined according to each land system type and each change driving factor, and the land utilization change simulation method comprises the following steps:

for each subarea, let c=1, call a random forest algorithm, and determine the suitability of the c-th land system type to be allocated to the subarea based on each of the variable driving factors;

And if c is smaller than N, letting c=c+1, continuing to execute the calling random forest algorithm, and determining the land suitability of the c-th land system type allocated to the subarea based on each of the variable driving factors, wherein N is the number of the land system types.

According to the land utilization change simulation method provided by the invention, the land service capacity of each land system type is calculated, and the method comprises the following steps:

determining a land system map of the target area based on the land system type of each subarea in the target area, and determining a land current map of the target area based on the land utilization current data of the target area, wherein the resolution of the land system map is higher than that of the land current map;

and calculating the land service capacity of each land system type according to the land system diagram and the land current situation diagram.

According to the land use change simulation method provided by the invention, the determining the competitive advantage of each subarea according to each land system type, each land service capability and each land service demand comprises the following steps:

inputting the land system type, the land service capacity and the land service demand corresponding to each subarea into a competitive advantage function for calculation to obtain the competitive advantage of the subarea;

The competitive advantage function is represented by the following formula (1):

（1）

wherein P_cmp _c,i,j Representing the competitive advantage of the jth land system type in the c-th sub-region in the ith internal iteration; CA (CA) _j,d Representing a j-th land system type providing a land service capability of a d-th land service; CA (CA) _u,d Represents the (u) th land systemThe system type provides land service capability of the d-th land service; the ith land system type is the land system type of the ith sub-area; parameter ineertia _d,i Representing the cumulative difference between the d-th land service and the service provided by each of the land system types at the end of the (i-1) th internal iteration.

According to the land use change simulation method provided by the invention, the conversion resistance and the neighborhood influence of each land system type are calculated according to the land use current situation data of the target area, and the method comprises the following steps:

based on the land utilization status data of the target area, evaluating the difficulty degree of converting each land system type into other land system types, and acquiring the conversion resistance of each land system type;

and calculating neighborhood influence of converting the land system type of each subarea into other land system types respectively based on the land utilization status data and the land system types of adjacent subareas of each subarea.

According to the land utilization change simulation method provided by the invention, the land conversion potential of each subarea is calculated according to each land suitability, each conversion resistance, each neighborhood influence and each competitive advantage, and the land utilization change simulation method comprises the following steps:

for each sub-area, if the type of the land system of the sub-area is the same as the type of the land system to be converted, taking the sum of the land suitability, the conversion resistance, the neighborhood influence and the competitive advantage corresponding to the sub-area as the land conversion potential; and if the land system type of the subarea is different from the land system type to be converted, taking the sum of the land suitability, the neighborhood influence and the competitive advantage corresponding to the subarea as the land conversion potential.

The invention also provides a land use change simulation device, which comprises:

the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is configured to acquire the land system type of each subarea in a target area, each variable driving factor corresponding to the target area and each land service demand;

a first determining module configured to determine a land suitability of each sub-region according to each of the land system types and each of the varying drive factors, and calculate a land service capacity of each of the land system types;

A second determination module configured to determine a competitive advantage of each sub-area based on each of the land system type, each of the land service capacity, and each of the land service demand;

a calculation module configured to calculate a conversion resistance and a neighborhood impact for each land system type based on land use present data for the target area; calculating the land transformation potential of each subarea according to the suitability of each land, the transformation resistance, the influence of each neighborhood and the competitive advantage;

and the simulation module is configured to input the land utilization current situation data, the land conversion potential, the land service capacity and the land service demand into an improved land utilization model to simulate land utilization change, so as to obtain a simulation result.

The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the land use change simulation method as described in any one of the above when executing the program.

The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a land use change simulation method as described in any of the above.

The invention also provides a computer program product comprising a computer program which when executed by a processor implements a land use change simulation method as described in any one of the above.

According to the land utilization change simulation method provided by the invention, the land system type of each subarea in the target area, each change driving factor corresponding to the target area and each land service demand are obtained; determining the land suitability of each subarea according to each land system type and each change driving factor, establishing complex and nonlinear relation between the land suitability and the driving factors, and calculating the land service capacity of each land system type; determining the competitive advantage of each subarea according to each land system type, each land service capability and each land service demand; calculating conversion resistance and neighborhood influence of each land system type according to the land utilization current situation data of the target area; calculating the land transformation potential of each subarea according to the suitability of each land, the transformation resistance, the influence of each neighborhood and the competitive advantage; and inputting the land utilization current situation data, the land conversion potential, the land service capacity and the land service demand into an improved land utilization model to simulate land utilization change to obtain a simulation result, and promoting the internal iteration of the land utilization model to realize supply and demand balance based on a stepwise iteration method so as to improve the efficiency and the reliability of the land utilization change simulation.

Drawings

In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a land use change simulation method provided by the invention;

FIG. 2 is a flow chart of an improved CLUMondo model iteration provided by the present invention;

FIG. 3 is a schematic diagram of a land use change simulation device provided by the invention;

fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In order to facilitate a clearer understanding of various embodiments of the present invention, some relevant background knowledge is first presented as follows.

The development and utilization degree of the land resource by human is continuously enhanced, and the development and utilization degree of the land resource is a main reason for the change of the earth surface layer, particularly the land surface layer. However, the problems of environmental risk and resource shortage caused by the reformation of the land by the human being threaten the survival and development of the human being, and how to reasonably utilize the land resource has become an important issue in the present era. Land utilization change simulation is an important component in land change science, and can infer future land space patterns according to the current or specific time development rules. By analyzing different future development scenario assumptions, the influence of different development scenario assumptions, plans, policies and the like on the land space pattern can be judged by using land utilization change simulation. Land utilization change simulation is also an important support for other research fields, such as important input of global environment change models and core focus of research fields of environmental management, biodiversity, carbon neutralization, and the like.

Over the past few decades, scientists have developed many different types of land use change simulation models to achieve future land use change simulations that are more scientific and meet the expected needs. Land use change simulation models can be divided into four types: the method comprises an empirical statistical model, a random model, an optimization model and a model based on a land dynamic mechanism, wherein the empirical statistical model establishes a quantitative relationship between land and possible external driving factors by using a regression method, and is widely applied to land change simulation research.

Among the empirical statistical models, the clirondo model is advantageous over most other empirical statistical models, and can support functions or services carried by the land as termination conditions for the simulation. In addition, the clutondo model supports establishing a many-to-many correspondence between land types and their services, which enables supply-demand balancing through iteration, so that services can be satisfied by the areas of the plurality of land types that provide them, rather than being limited to a particular certain land type.

However, the CLUMondo model uses a logic regression method to establish a linear relationship between the suitability parameter and the biophysical and socioeconomic factors, that is, the CLUMondo model uses a linear model when establishing a relationship between the suitability parameter and the variable driving factor, and the internal operation mechanism is not enough to fit the actual situation, so that the accuracy of the simulation result is affected. And the original internal operating mechanism of the clirondo model causes a large change in each iteration, such a framework is beneficial to the rapid convergence of demand and supply in early iterations, but may be detrimental to model convergence when the difference between demand and supply is small.

Therefore, the invention provides a land utilization change simulation method, which is characterized by obtaining the land system type of each subarea in a target area, each change driving factor corresponding to the target area and each land service demand; determining the land suitability of each subarea according to each land system type and each change driving factor, establishing complex and nonlinear relation between the land suitability and the driving factors, and calculating the land service capacity of each land system type; determining the competitive advantage of each subarea according to each land system type, each land service capability and each land service demand; calculating conversion resistance and neighborhood influence of each land system type according to the land utilization current situation data of the target area; calculating the land transformation potential of each subarea according to the suitability of each land, the transformation resistance, the influence of each neighborhood and the competitive advantage; and inputting the land utilization current situation data, the land conversion potential, the land service capacity and the land service demand into an improved land utilization model to simulate land utilization change to obtain a simulation result, and promoting the internal iteration of the land utilization model to realize supply and demand balance based on a stepwise iteration method so as to improve the efficiency and the reliability of the land utilization change simulation.

The land use change simulation method and device provided by the invention are described below with reference to fig. 1.

Fig. 1 is a schematic flow chart of a land use change simulation method provided by the invention, and referring to fig. 1, the method includes steps 101 to 105, wherein:

step 101: and acquiring the land system type of each subarea in the target area, each change driving factor corresponding to the target area and each land service demand.

It should be noted that the execution subject of the present invention may be any electronic device simulating land utilization change, for example, any one of a smart phone, a smart watch, a desktop computer, a laptop computer, and the like.

Specifically, the target area refers to a land, i.e., a research area, where a land use change simulation is required. A sub-region refers to a grid that divides a target region. The land system type, i.e. the land system data, refers to the type of sub-area in both land use and density dimensions. The change driving factors comprise land utilization change driving factors and corresponding numerical values, are the driving of the socioeconomic and biophysical data on land utilization change, and comprise eight types, namely socioeconomic factors, soil factors, accessibility factors, agricultural and vegetation factors, topography factors, climate factors, livestock factors and duty factors. Land service refers to the use service provided by land and includes four types of forest, grassland/pasture, shrub and cultivated land. The individual land service demand refers to a supply amount, such as a required area, for each land service.

In practical application, the target area can be determined first, and then the target area is divided into a plurality of subareas, so that the land system type of each subarea, each change driving factor corresponding to the target area and each land service demand are obtained.

Various methods for acquiring the land system type of each sub-area, each variable driving factor corresponding to the target area and each land service demand exist, for example, a user uploads the land system type of each sub-area, each variable driving factor corresponding to the target area and each land service demand through an uploading page, and accordingly, the execution main body acquires the land system type of each sub-area, each variable driving factor corresponding to the target area and each land service demand; for another example, the execution body receives the land use change simulation instruction or the acquisition instruction, and correspondingly, the execution body acquires the land system type of each sub-region, each change driving factor corresponding to the target region and each land service demand from the storage region pointed by the land use change simulation instruction or the acquisition instruction. The invention is not limited in this regard.

In addition, the land service demand can be acquired through the global change analysis model, namely, the land service demand in future scenes is acquired from the global simulation result of the global change analysis model.

Illustratively, land use in future scenarios is obtained from the simulation results of the global change analysis model v 5.3. The global change analysis model v5.3 outputs 70 land types, and divides the 70 land types into four land services of woodland, grassland/pasture, shrub land and cultivated land, so as to obtain the required quantity of each river basin for the four land services in future situations, namely the required quantity of each land service.

Step 102: and determining the land suitability of each subarea according to each land system type and each change driving factor, and calculating the land service capacity of each land system type.

Specifically, land adaptability refers to the suitability of a current land system type to be assigned to a current sub-area. Land service capability refers to the capability of the current land system type to provide a particular land service, featuring a land service that is any land service.

In practical application, on the basis of obtaining each land system type and each change driving factor, further, according to each land system type and each change driving factor, a random forest algorithm is used to calculate the suitability of each land system type allocated to a certain subarea, and the land suitability is obtained. Meanwhile, based on the target area, the land service capacity of each land system type is calculated.

It should be noted that, to facilitate debug, the original format of all the variable driving factors may be converted into ASCII format.

Step 103: and determining the competitive advantage of each subarea according to each land system type, each land service capability and each land service demand.

In particular, the competitive advantage refers to the advantage that the current sub-area possesses over other sub-areas in providing the land service.

In practical application, on the basis of obtaining the types of the land systems, the land service capacities and the land service demands, the types of the land systems, the land service capacities and the land service demands can be further calculated according to a set competitive advantage algorithm, so that the competitive advantage of each subarea is obtained.

Step 104: calculating conversion resistance and neighborhood influence of each land system type according to the land utilization current situation data of the target area; and calculating land conversion potential of each sub-region based on each of the land suitability, each of the conversion resistances, each of the neighborhood effects, and each of the competitive advantages.

Specifically, the land use status data characterizes the land use condition of the current target area, and the land use type refers to the type of the subarea in the land use dimension, and the land use type includes ten types, namely cultivated land, woodland, grassland, shrub land, wetland, water body, moss, artificial earth surface, bare land and snow/ice. The conversion resistance judges whether each land system type can be converted into other land system types or not through the land system data, and the conversion difficulty degree is high. The neighborhood influence considers the influence of the sub-region by the surrounding sub-region (adjacent sub-region), by which is meant the determination that the land system type of the surrounding sub-region influences the land system type of the sub-region.

In practical application, the conversion resistance of each land system type can be calculated by taking land utilization current data as input according to a set conversion resistance algorithm, and the neighborhood influence of each land system type can be calculated by taking land utilization current data as input according to a set neighborhood influence algorithm.

On the basis of obtaining the suitability of each land, the conversion resistance, the influence of each neighborhood and the competitive advantage, further, the suitability of each land, the conversion resistance, the neighborhood shadow and the competitive advantage can be input into an improved CLUMondo model, and the land conversion potential is calculated; the land transformation potential of each subarea can be calculated by taking the suitability of each land, each transformation resistance, each neighborhood shadow and each competitive advantage as inputs according to a set calculation rule.

Step 105: and inputting the land utilization current data, the land conversion potential, the land service capacity and the land service demand into an improved land utilization model to simulate land utilization change, so as to obtain a simulation result.

In particular, the land use model, i.e. the land use variation simulation model, is preferably an improved land use model, which is an improved clutondo model.

In practical application, on the basis of obtaining each land conversion potential, each land service capacity and each land service demand, further, inputting each land system data, each land service capacity, each land service demand and each land conversion potential into an improved dump model to perform land utilization change simulation, and performing nested iteration on the land system data, each land service capacity, each land service demand and each land conversion potential by using the improved land utilization model to obtain a simulation result of the land utilization change simulation.

According to the land utilization change simulation method provided by the invention, the land system type of each subarea in the target area, each change driving factor corresponding to the target area and each land service demand are obtained; determining the land suitability of each subarea according to each land system type and each change driving factor, establishing complex and nonlinear relation between the land suitability and the driving factors, and calculating the land service capacity of each land system type; determining the competitive advantage of each subarea according to each land system type, each land service capability and each land service demand; calculating conversion resistance and neighborhood influence of each land system type according to the land utilization current situation data of the target area; calculating the land transformation potential of each subarea according to the suitability of each land, the transformation resistance, the influence of each neighborhood and the competitive advantage; the land system data, the land conversion potential, the land service capacity and the land service demand are input into an improved land utilization model to simulate land utilization change, a simulation result is obtained, the internal iteration of the land utilization model can be promoted to realize supply and demand balance based on a stepwise iteration method, and further the efficiency and the reliability of the land utilization change simulation are improved.

In one or more optional embodiments of the invention, the acquiring a land system type of each sub-region in the target region includes:

In practical application, the present data of the land use of the target area is acquired first, then, for each sub-area, the present data of the sub-land use corresponding to the sub-area is identified from the present data of the land use, and the type of the land use of the sub-area is identified according to the present data of the sub-land use: according to the current status data of the sub-land utilization, determining the land area corresponding to each initial land utilization type in the sub-area, and determining the initial land utilization type corresponding to the maximum land area as the target land utilization type of the sub-area, namely, the target land utilization type of the sub-area is the initial land utilization type with the highest area occupation ratio or the largest area in the range of the sub-area.

Further, for each land use type, subdividing the current land use type from the density dimension to obtain the land system type of each subarea in the target area: determining a target subarea belonging to each land utilization type according to each land utilization type, determining the ratio of a target area corresponding to the land utilization type in each target subarea to the target subarea, namely the area ratio, traversing all the target subareas to obtain each area ratio corresponding to the land utilization type, performing density division on the area ratio, dividing the land utilization type into a plurality of land system types, and further obtaining the land system type of each target subarea. Traversing all land utilization types to obtain the land system type of each target subarea in the target area. Thus, the land system type can be determined based on the area ratio, and the accuracy and reliability of the land system type can be improved.

Illustratively, the target area is taken as an area a, and the latest Globeland30 land utilization data and the boundary data of the area a issued by the data center are utilized to cut off the data of the area a, namely the land utilization status data, wherein Globeland30 is 30 m space resolution global earth surface coverage data. Globeland30 land utilization data was aggregated from 30 meters resolution to 990 meters resolution. For each 990 x 990 meter large grid, i.e. sub-region, the land use type is defined as the land use type with the highest area occupation ratio in the range. For each land use type, the actual area proportion, i.e. the area duty ratio, of the land use type in a large grid with all types in the target area being the land use type is counted. For example, a large grid of the forest type, in which the area of the forest is 900×900, the duty cycle is 900×900/(990×990) =100/121. And determining the area ratio of each subarea corresponding to the land utilization type, dividing the land utilization type into a plurality of land system types according to the density division of the area ratio, and further obtaining the land system type of each target subarea, namely further dividing the land utilization type into three density levels of low, medium and high to obtain three land system types, for example, land utilization type is cultivated land, and the corresponding three land system types are respectively low-density cultivated land, medium-density cultivated land and high-density cultivated land, so as to further obtain the land system type of each target subarea.

In one or more optional embodiments of the present invention, the determining a land system type of each sub-area in the target area according to each land utilization type and each area ratio may be implemented as follows:

Specifically, the natural breakpoint algorithm may identify the separation of the classes, may group the similarity values most appropriately, and may maximize the differences between the classes; the elements will be divided into classes for which their boundaries, i.e. thresholds, will be set at locations where the differences in data values are relatively large.

In practical application, for each land use type, a natural breakpoint algorithm is used to divide all area ratios corresponding to the land use type into three groups, namely a first threshold value and a second threshold value are determined.

In the case where the first threshold is greater than the second threshold: determining the density level of the land utilization type of the subarea with the area ratio larger than the first threshold value as a high density level, and further determining the land system type of the subarea, wherein the land system type of the subarea is high-density wetland if the land utilization type is wetland; determining a density level of a land use type of a subarea with an area ratio greater than a second threshold value and less than or equal to a first threshold value as a medium density level, and further determining a land system type of the subarea, for example, a land use type is a grassland, and the land system type of the subarea is a medium density grassland; and determining the density level of the land utilization type of the subarea with the area ratio smaller than or equal to the second threshold value as a low density level, and further determining the land system type of the subarea, for example, if the land utilization type is a water body, the land system type of the subarea is a low density water body.

In the case where the first threshold is less than the second threshold: determining the density level of the land utilization type of the subarea with the area ratio larger than a second threshold value as a high density level, and further determining the land system type of the subarea; determining the density level of the land utilization type of the subarea with the area ratio larger than a first threshold value and smaller than or equal to a second threshold value as a medium density level, and further determining the land system type of the subarea; and determining the density level of the land utilization type of the subarea with the area ratio smaller than or equal to the first threshold value as a low density level, and further determining the land system type of the subarea.

Specifically, for each land use type, calling a natural breakpoint algorithm for the area occupation ratio of each subarea corresponding to the land use type, and determining a first threshold value and a second threshold value, wherein the specific implementation process is as follows:

calculating a first difference of each area occupation ratio, and determining all grouping results for dividing each area occupation ratio into three groups;

calculating a second variance of each group in the group result aiming at each group result, and adding the second variances to obtain a variance sum of the group results;

calculating the variance fitting goodness of each grouping result according to the variance sum of each grouping result;

and determining a first threshold and a second threshold according to the maximum variance fitting goodness-of-fit corresponding grouping.

Specifically, variance refers to the sum of squares of deviations of the average values.

In practical application, firstly, according to

Calculating a first variance of each area ratio corresponding to the current land utilization type, wherein SDAM is a firstVariance x _i Is the i-th area ratio, namely the actual area ratio value of the land use type in the i-th sub-area corresponding to the current land use type, +.>

And the average value of all area occupied ratios corresponding to the land utilization type is used.

Then, for each possible grouping combination, i.e. grouping result (three groups are included in each grouping result), according to

Calculating the sum of variances of each grouping result and finding the grouping result with the smallest sum of variances, wherein SDCM is the sum of variances, x _ij Is the i-th area ratio in the j-th group,>

is the average of all area ratios of group j,

and is the second variance. Since land utilization types are divided into three density levels, j is a positive integer of 3 or less.

Further, the variance fitting goodness of each grouping result is calculated according to gvf= (SDAM-SDCM)/SDAM, wherein GVF is the variance fitting goodness. The GVF can range from 1 (best packet) to 0 (worst packet). Finding the highest grouping result of GVF, determining two thresholds of density level, namely a first threshold and a second threshold.

Therefore, by adopting a natural breakpoint algorithm, the reliability and the efficiency of the first threshold value and the second threshold value can be improved, and the reliability and the efficiency of determining the type of the land system are further improved.

It should be noted that after determining the land system type of each sub-area, the spatial resolution needs to be adjusted to match the spatial resolution of most of the variable driving factors. In the above example, after determining the land system type of each sub-area, resampling the spatial resolution from 990m x 990m to 1km x 1km, wherein the resampling uses the nearest resampling method, and each grid attribute value of 1km x 1km is the grid attribute value of the 990m x 990m data sample center nearest to the grid center.

In one or more alternative embodiments of the present invention, the determining the land suitability of each sub-area according to each land system type and each variable driving factor may be implemented as follows:

In practical application, a random forest method is used for calculating the land suitability: p_loc _c，j This parameter indicates the suitability of the jth land system type to be assigned to the c-th sub-area. The calculation method is as follows: first, the variable driving factor (value) on the sub-area c is prepared: (X) _1,c ,X _2,c ,…,X _m,c ) Where m represents the number of varying driving factors and c represents the c-th sub-region. And (3) establishing a random forest model for each land system type j, wherein a response variable used in training the random forest model is 0,1 binary, and when the current land system type of the subarea c is the jth land system type, the value of the response variable corresponding to the subarea c is determined to be 1, otherwise, the value of the response variable is 0. The varying driving factors and response variables over all sub-regions together constitute a sample set. Establishing the suitability of the driving factors and the land by training a random forest model: p_loc _c，j Relationship between them. Further, all sub-areas and all land system types are traversed. Thus, by coupling a random forest algorithm, a complex and nonlinear relation between the land suitability and the change driving factor is established, the relation between the change driving factor and the land suitability can be reflected more accurately, and the improvement of land utilization change is facilitatedThe reliability of the simulation is improved.

Alternatively, to improve the accuracy of land suitability, all random forest models are provided with 200 trees.

In one or more alternative embodiments of the invention, the computing the land service capacity of each of the land system types includes:

In practical application, a land system diagram of a target area is obtained based on the land system type of each subarea in the target area, and a land current diagram of the target area is obtained based on the land utilization current data of the target area. Then the land system diagram and the land current situation diagram are overlapped according to CA _j,d =S _j,d /S _j Calculating the land service capacity of each land system type, wherein CA _j,d Is the land service capability of the jth land system type relative to the jth land service (one of forest, grassland/pasture, shrub, cultivated land), S _j Is the total area of the jth land system type in the target area S _j,d Is the supply of the d-th land service on the j-th land system type. Therefore, the land service capacity of the land system type can be calculated rapidly, the accuracy of the land service capacity is improved, and the accuracy of land utilization change simulation is improved.

Illustratively, 30 land system types (30 land system types are obtained after 10 land utilization types are divided into three density levels) of the target area are calculated by superimposing a land system map (resolution 990 m) and a Globeland30 map (resolution 30m, i.e., a land current map), respectively _j,d 。

In addition, S _j,d The calculation method of (2) meets the following conditions that the land utilization type areas are summed: (1) The 30m x 30m sub-region overlaps with the j-th land system type on the land system map; (2) The land use type of 30m x 30m can provide the d-th land service.

In one or more alternative embodiments of the present invention, the determining the competitive advantage of each sub-area according to each land system type, each land service capability, and each land service demand may be implemented as follows:

the competitive advantage function is represented by the following formula (1):

（1）

wherein P_cmp _c,i,j Representing the competitive advantage of the jth land system type in the c-th sub-region in the ith internal iteration; CA (CA) _j,d Representing a j-th land system type providing a land service capability of a d-th land service; CA (CA) _u,d A land service capability indicating that the u-th land system type provides a d-th land service; the ith land system type is the land system type of the ith sub-area, that is, the land system current status type; parameter ineertia _d,i Representing the cumulative difference between the d-th land service and the service provided by each of the land system types at the end of the (i-1) th internal iteration is an adaptive inertial mechanism.

In practical application, on the basis of obtaining the types, the service capacities and the service demand amounts of all the land systems, the types, the service capacities and the service demand amounts of all the land systems are substituted into the competitive advantage function to calculate, so that the competitive advantage of the subareas is obtained. Therefore, the competitive advantage can be rapidly and accurately determined, and the efficiency and the reliability of land utilization change simulation are improved.

Wherein the inertial mechanism is ineertia _d,i Is represented by the following formula (2):

（2）

wherein Demand _d Representing the demand for the d-th land service, namely the d-th land service demand; supply _d,i-1 Representing the supply of the type of land system to the d-th land service at the end of the (i-1) th internal iteration. Is a speed parameter, expressed by the following formula (3):

（3）

in the formula (2) and the formula (3), ineertia _d,i A cumulative difference between the preset d-th land service and the service provided by the land system type at the end of the (i-1) th internal iteration; it is subjected to a speed parameter

Is controlled by (a);

triggered by a random Seed and increasing with increasing number of internal iterations, the initial values of the parameters Seed and Step are preferably 1 and 0.001, respectively.

In one or more optional embodiments of the present invention, the calculating the conversion resistance and the neighborhood impact of each land system type according to the land utilization status data of the target area may be implemented as follows:

In practical application, based on the land use status data, the difficulty level of converting the jth land system type into other land system types can be estimated according to a set estimation strategy, and the conversion resistance of the jth land system type can be obtained

Wherein j is a positive integer of not more than 30. Conversion resistance->

Is in the range of [0,1 ]]. A conversion resistance value of 0 for the land system type indicates that it is very easy to convert, and a value of 1 indicates that conversion is prohibited.

In addition, it is necessary to calculate the possibility of converting the land system type of the c-th sub-region into the j-th land system type, that is, the conversion resistance, based on the land system data in combination with the land system type of each neighborhood (adjacent sub-region) of the c-th sub-region

。/>

The greater the score, the greater the likelihood that it will be homoginized into the j-th land system type.

In one or more alternative embodiments of the present invention, the calculating the land transformation potential of each sub-region according to each of the land suitability, each of the transformation resistances, each of the neighborhood effects, and each of the competitive advantages may be implemented as follows:

Specifically, the to-be-converted land system type refers to a land system type after the current land system type is converted.

In practical application, aiming at the c-th sub-area, the type of the land system to be converted is the j-th land system type, and if the type of the land system of the c-th sub-area is the j-th land system type, the land conversion potential is realized

Is equal to the sum of land suitability, conversion resistance, neighborhood impact and competitive advantage; if the current soil status type of the c-th sub-area is not the j-th soil system type, then the soil conversion potential is +.>

Equal to the sum of land suitability, neighborhood impact and competitive advantage.

In practical application, referring to fig. 2, fig. 2 is a schematic flow chart of an improved CLUMondo model iteration provided in the present invention: the improved clutondo model is input with each land suitability, future land demand (each land service demand), land service capacity, competitive advantage, and the like, for performing a land change simulation based on the balance of the land supply and demand services. In the iteration process of the improved CLUMondo model, finer iterations are introduced, nested in each external iteration. A finer iteration will be triggered if the following conditions are met: all internal iterations of an external iteration are performed and the improved clutondo model iterates out of the land system that cannot meet future land service requirements, i.e. the model does not converge. Finer iterations first determine the highest land conversion potential in all the sub-regions that can be changed and in all the possible land system types

I.e. +. >

I.e.)>

Wherein->

Representing the land conversion potential of any sub-region c into any land system type k; q is a counting variable, and the value range is 1 to n (consistent with the number of subareas) and is used for traversing all subareas in one iteration; k is the kth land system type. Then, it will->

The land system type is allocated to +.>

The sub-region then checks whether the modified CLUMondo model converges. If convergence is not reached +.>

The sub-region will be removed from further consideration and the previous steps repeated. The calculation is then iterated until the improved clirondo model matches the simulated land amount to the land demand, or all sub-areas try to go through the land system conversion in finer iterations. Therefore, by improving the internal operation mechanism of the CLUMondo model and using improved competitive advantage parameters and finer iteration, the land utilization model can be finely adjusted, the iteration of the land utilization model can be promoted to reach supply and demand balance, and the efficiency and the reliability of land utilization change simulation are improved.

More specifically, see fig. 2: the modified CLUMondo model simulates the change in land through two nested iterations: the outer iteration t may cover the entire simulation period; when the simulation result of land change at a certain time in the middle needs to be output, the external iteration t can also only cover part of the simulation time period. Wherein T ranges from 1 to T. Each outer iteration contains a large number of inner iterations, preferably 20000 inner iterations i, i.e. i ranging from 1 to 20000.

In each internal iteration, the modified CLUMondo model determines according to equation (4) whether or not the land system type of each sub-area changes, to which land system type:

（4）

in equation (4), c represents the c-th sub-region, t represents the t-th outer iteration, and i represents the i-th inner iteration of the t-th outer iteration.

And the land system type of the ith sub-area after the ith internal iteration in the ith external iteration is ended is represented. />

Is the land switching potential, and represents the probability that the land system type of the c-th sub-area switches (or continues to hold) to the j-th (1.ltoreq.j.ltoreq.n, n being the total number of land system types) land system type. Phi is an area (e.g., an ecological protection area) in the study area (target area) where the occurrence of land system changes is prohibited. />

(T (c, T, 0)) for counting, recording the land system type of the c-th sub-area and the land system type of the c-th sub-area in the land system map, i.e. & lt & gt>

Consistent number of external iterations. />

(T (c, T, 0)) represents the minimum number of outer iterations that the land system type of the c-th sub-zone should remain unchanged, and is a non-negative integer. />

Control whether to allow land system type +. >

And converting into a land system type k or a kth land system type.

I.e. to calculate'

", first judging whether c E phi is true, if yes, then ++>

=/>

If not, continue to judge +.>

(T(c, t, 0))＜/>

(T (c, T, 0)) is true, if so, +.>

=/>

If not, judge->

Whether or not it is true, if not, then->

=/>

If yes, then

= />

. Further judging whether the improved CLUMondo model is converged or not, if so, interrupting or stopping; if not, let i=i+1, continue the internal iteration. If after i=20000 the improved clamondo model is still not converging, a finer iteration is started, calculating +.for i (from 20001 to 20001+n)>

Let->

Wherein->

And->

Is to make land transformation potential +.>

The combination of parameters taking the maximum value, i.e. +.>

The sub-region is converted into +.>

The land transfer potential of the land system type is greatest. Judging whether the improved CLUMondo model is converged or not, if so, interrupting or stopping; if not, let->

The finer iteration continues.

Determining

Key to (2) land conversion potential +.>

Which is obtained by adding four parameters affecting the land change in the model, the four parameters are land suitability +.>

Conversion resistance->

Neighborhood influence>

And competitive advantage->

. The calculation formula of the land transformation potential is represented by the following formula (5):

（5）

The conversion resistance judges whether each land system type can be converted into other land system types and the difficulty degree of conversion through historical land data, and each land system type corresponds to 1 elastic coefficient and has the range of [0,1]. A conversion resistance value of 0 for a land system type indicates that it is very easy to convert, and a value of 1 indicates that conversion is prohibited.

The neighborhood impact considers not only each block itself, but also the impact of surrounding blocks of the block, wherein the impact refers to the judgment that the surrounding block type affects the conversion of the center block type,

the greater the score, the greater the likelihood that it will be homogenized to the land system type j.

And inputting the land suitability, the neighborhood influence, the conversion resistance and the improved competitive advantage calculation result into the improved CLUMondo model to calculate land conversion potential. After the land system of the initial year is input, the model configures the land system based on the land conversion potential, and the total land service supply is calculated according to the area of each land system type and the land service supply capacity after each iteration. And when the difference between the total supply quantity and the total demand of the predicted year is within a threshold value, finishing simulation, otherwise, adjusting the competitive advantage of the model according to the supply condition of the land service demand and the competitive capacity of the land system type, recalculating the land conversion potential, configuring a land system for a new iteration according to the new land conversion potential, and judging whether the land after the iteration meets the land service demand of the predicted year again until the land system iterated out meets the future land service demand.

According to the land utilization change simulation method provided by the invention, the random forest algorithm is used for replacing the binary Logit regression algorithm, the relation between the change driving factor of land utilization change and the current situation of a land system is established, the land suitability is calculated, the reference of the occurrence probability of each land system is given according to the established random forest algorithm when the land utilization model simulates land change, and the complex nonlinear relation between the change driving factor and the land suitability can be reflected more accurately. The improved competitive parameters and finer iterations are used to allow the model to be fine tuned to achieve a supply-demand balance.

The land use change simulation device provided by the invention is described below, and the land use change simulation device described below and the land use change simulation method described above can be referred to correspondingly.

Fig. 3 is a schematic structural diagram of a land use change simulation device provided by the present invention, and as shown in fig. 3, the land use change simulation device 300 includes: an acquisition module 301, a first determination module 302, a second determination module 303, a calculation module 304, and a simulation module 305, wherein:

an obtaining module 301, configured to obtain a land system type of each sub-region in a target region, each variable driving factor corresponding to the target region, and each land service demand;

A first determining module 302 configured to determine a land suitability of each sub-area according to each of the land system types and each of the varying drive factors, and calculate a land service capacity of each of the land system types;

a second determining module 303 configured to determine a competitive advantage of each sub-area according to each of the land system types, each of the land service capacities, and each of the land service demands;

a calculation module 304 configured to calculate a conversion resistance and a neighborhood impact for each land system type based on land utilization status data for the target area; calculating the land transformation potential of each subarea according to the suitability of each land, the transformation resistance, the influence of each neighborhood and the competitive advantage;

the simulation module 305 is configured to input the land use status data, each of the land conversion potentials, each of the land service capacities, and each of the land service demands into an improved land use model to perform land use change simulation, and obtain a simulation result.

According to the land utilization change simulation device provided by the invention, the land system type of each subarea in the target area, each change driving factor corresponding to the target area and each land service demand are obtained; determining the land suitability of each subarea according to each land system type and each change driving factor, establishing complex and nonlinear relation between the land suitability and the driving factors, and calculating the land service capacity of each land system type; determining the competitive advantage of each subarea according to each land system type, each land service capability and each land service demand; calculating conversion resistance and neighborhood influence of each land system type according to the land utilization current situation data of the target area; calculating the land transformation potential of each subarea according to the suitability of each land, the transformation resistance, the influence of each neighborhood and the competitive advantage; and inputting the land utilization current situation data, the land conversion potential, the land service capacity and the land service demand into an improved land utilization model to simulate land utilization change to obtain a simulation result, and promoting the internal iteration of the land utilization model to realize supply and demand balance based on a stepwise iteration method so as to improve the efficiency and the reliability of the land utilization change simulation.

Optionally, the obtaining module 301 is further configured to:

Optionally, the first determining module 302 is further configured to:

Optionally, the second determining module 303 is further configured to:

the competitive advantage function is represented by the following formula (1):

（1）

wherein P_cmp _c,i,j Representing the competitive advantage of the jth land system type in the c-th sub-region in the ith internal iteration; CA (CA) _j,d Representing a j-th land system type providing a land service capability of a first land service; CA (CA) _u,d A land service capability indicating that the u-th land system type provides a d-th land service; the ith land system type is the land system type of the ith sub-area; parameter ineertia _d,i Representing the cumulative difference between the d-th land service and the service provided by each of the land system types at the end of the (i-1) th internal iteration.

Optionally, the computing module 304 is further configured to:

Fig. 4 illustrates a physical schematic diagram of an electronic device, as shown in fig. 4, which may include: processor 410, communication interface (Communications Interface) 420, memory 430 and communication bus 440, wherein processor 410, communication interface 420 and memory 430 communicate with each other via communication bus 440. Processor 410 may invoke logic instructions in memory 430 to perform a land use change simulation method.

Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In another aspect, the present invention also provides a computer program product comprising a computer program storable on a non-transitory computer readable storage medium, the computer program, when executed by a processor, being capable of performing the land use change simulation method provided by the methods described above.

In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the land use change simulation method provided by the above methods.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A land use change simulation method, characterized by comprising:

2. The land use variation simulation method according to claim 1, wherein the acquiring the land system type of each sub-area in the target area comprises:

3. The land use variation simulation method according to claim 2, wherein said determining a land system type of each sub-region in said target region based on each of said land use types and each of said area ratios comprises:

4. The land use variation simulation method according to claim 1, wherein said determining the land suitability of each sub-area based on each of the land system types and each of the variation driving factors comprises:

5. The land use variation simulation method according to claim 1, wherein said calculating land service capacity of each of said land system types comprises:

6. The land use variation simulation method according to claim 1, wherein said determining the competitive advantage of each sub-area based on each of said land system type, each of said land service capacity, and each of said land service demand comprises:

the competitive advantage function is represented by the following formula (1):

（1）

Wherein P_cmp _c,i,j Representing the competitive advantage of the jth land system type in the c-th sub-region in the ith internal iteration; CA (CA) _j,d Representing a j-th land system type providing a land service capability of a d-th land service; CA (CA) _u,d A land service capability indicating that the u-th land system type provides a d-th land service; the ith land system type is the land system type of the ith sub-area; parameter ineertia _d,i Representing the cumulative difference between the d-th land service and the service provided by each of the land system types at the end of the (i-1) th internal iteration.

7. The land use variation simulation method according to claim 1, wherein said calculating the conversion resistance and neighborhood impact of each land system type based on the land use present data of the target area comprises:

8. The land use variation simulation method according to claim 1, wherein said calculating land conversion potential of each sub-area based on each of said land suitability, each of said conversion resistances, each of said neighborhood effects, and each of said competitive advantages comprises:

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the land use change simulation method according to any one of claims 1 to 8 when the program is executed by the processor.

10. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the land use change simulation method according to any one of claims 1 to 8.