Land space planning and fine management method based on big data

Abstract

The invention discloses a land space planning and fine management method based on big data, and relates to the technical field of land planning. A random forest model is adopted to dynamically adjust a weight factor, and factors such as land availability, ecological sensitivity and infrastructure bearing capacity are combined, so that the land space planning and fine management can be realized. The resource bearing capacity is comprehensively quantified, high, medium and low bearing capacity areas are divided, then ecological protection, industrial layout and public service coverage references are combined, an analytic hierarchy process (AHP) is adopted to calculate suitability scores of ecology, production and life, and accurate space division is realized through a rasterization method. Finally, the performability and adaptability of planning are enhanced through guided linear planning and a dynamic feedback mechanism, space conflicts are optimized through scene simulation, model parameters are continuously adjusted through remote sensing and IoT monitoring data, dynamic synchronization of planning and reality is achieved, and the problems that in traditional planning, static performance is high, and adjustment lags are effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of land planning technology, and in particular to a land space planning and refined management method based on big data. Background Art

[0002] With the rapid advancement of urbanization, the intensification of resource and environmental pressures, and the diversified development of demands, the existing land space planning schemes have gradually revealed their limitations; most of the existing schemes rely on static assessment and planning methods, and complete the preliminary layout of resources, environment and population based on one-time data collection, resulting in poor results in dynamic applications.

[0003] Current land spatial planning is mostly based on data accumulated in the past few years and evaluated in combination with empirical models. However, with the rapid economic development and intensified population mobility, land use, resource demand and environmental carrying capacity have changed significantly. In rapidly developing suburban areas, population density and infrastructure demand may increase significantly in the short term, but existing plans often underestimate the carrying capacity of these areas, which in turn leads to a lag in resource allocation and spatial planning. In addition, the spatial division accuracy in existing management methods is insufficient, which limits the actual implementation effect of the plan. Traditional methods mostly divide space based on administrative units, such as counties, cities or districts. However, this division scale is too large to reflect the functional differences within the region. The same county may contain multiple functional areas such as farmland, industrial parks and ecological protection areas, but these functions are treated uniformly within a unified administrative boundary, resulting in inefficient resource allocation and management measures, exacerbating land use conflicts and resource waste.

[0004] In order to solve the above problems, some traditional solutions increase the complexity of planning indicators and evaluation models to improve accuracy. For example, the dual evaluation method, namely resource and environmental carrying capacity evaluation and land space development suitability evaluation, has become an important means of planning in recent years. However, the dynamic and real-time data of this solution is still difficult to guarantee, and the evaluation results still rely on static data and assumptions at specific time points. At the same time, in the division of multi-functional space, it is still difficult to scientifically and reasonably balance the contradiction between ecological protection and economic development. Therefore, a land space planning and refined management method based on big data is urgently needed to solve such problems. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides a national land space planning and refined management method based on big data to solve the problem that the existing land space planning methods still have significant deficiencies in coping with the dynamic changes in resource carrying capacity and multi-functional space division, which limits the operability and adaptability of the planning.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: The embodiment of the present invention provides a land space planning and refined management method based on big data, which includes: Step S1, acquiring planning data in real time, including land resources, ecological environment and socio-economic data, and constructing a resource status data set based on the planning data; Step S2, based on the resource status data set, construct a multi-objective weight model to evaluate resource carrying capacity, and generate a regional resource carrying capacity classification map, in which high, medium and low carrying capacity areas are divided; Step S3, combining the resource classification map and demand benchmark, dividing the three life spaces, including ecology, production and life; Step S4, using a linear programming LP model to perform multi-scenario simulation and output a partitioning scheme that meets local goals; In step S5, monitoring data including land use and environmental changes are captured through remote sensing images and IoT devices, the difference between the monitoring data and the partitioning scheme generated in step S4 is evaluated, the deviation results are updated to the data set of step S1, and the weight factors and parameters of the comprehensive suitability evaluation model are optimized.

[0008] As a preferred solution of the national land space planning and refined management method based on big data described in the present invention, in step S1, the land resources, ecological environment and socio-economic data are obtained in the following manner: Obtain land resource data through remote sensing image platforms, including forest land, cultivated land and building land cover types, Obtain real-time ecological and environmental data through meteorological monitoring stations and environmental monitoring IoT devices, including water resource distribution, air quality, and climate indicators. Socioeconomic data, including population density, intensity of economic activities, layout of transportation networks, and distribution of public facilities, are obtained through statistical yearbooks and mobile signaling data.

[0009] As a preferred solution of the national land space planning and refined management method based on big data described in the present invention, the weight factor of the multi-objective weight model is dynamically adjusted according to the historical planning effectiveness data using a random forest regression model.

[0010] As a preferred solution of the national land space planning and refined management method based on big data described in the present invention, in step S2, the multi-objective weight model integrates three dimensions of resource carrying capacity, including: Land resource carrying factor: land availability, Ecological environment carrying factors: environmental sensitivity, Socioeconomic carrying factors: population pressure and infrastructure.

[0011] As a preferred solution of the national land space planning and refined management method based on big data described in the present invention, the steps of constructing a multi-objective weight model to evaluate resource carrying capacity based on the resource status data set and generating a regional resource carrying capacity classification map are as follows: The weighted linear model is used to construct the resource carrying capacity score, and the calculation formula is: , in, For Region The comprehensive score of resource carrying capacity is For Region The land resource carrying factor represents the proportion of available land. For Region The ecological environment carrying factor indicates ecological sensitivity. For Region The socioeconomic carrying factor represents population pressure and infrastructure level. are the weight factors of the three dimensions of land resources, ecological environment and social economy. Each factor is calculated based on the resource status data, and the calculation formula is: , in, For Region of available land area, For Region of total land area, , is the ecological sensitivity coefficient, ranging from 0 to 1, with 1 indicating extremely high sensitivity. Ecologically sensitive area The area of is the number of ecologically sensitive areas, , in, For Region The total population of For Region The carrying capacity of infrastructure; The threshold classification method The high, medium and low bearing capacity areas are divided as follows: High load-bearing capacity , The medium bearing capacity is , Low load capacity , in, For all regions The average value of For all regions The standard deviation of the generated classification map represents the distribution of regional carrying capacity in the form of a spatial grid; In step S2, the weight factor is dynamically adjusted as follows: Assume that there is a historical plan Regional samples, each sample records the actual planning results and the corresponding resource factor , define the model training objective as: ,in, Planning areas for history The actual planning results is the random forest regression model, is the random error term, The importance of each factor is calculated using the random forest model. , Factor The weight factor update formula for the relative importance of planning results is: , in, Factor Dynamic weight, substitute the dynamically adjusted weight factor into the formula In the process, the resource carrying capacity score is recalculated and the classification diagram is updated.

[0012] As a preferred solution of the national land space planning and refined management method based on big data described in the present invention, in step S3, the demand benchmark includes: Ecological needs: define ecological suitability areas based on ecological red lines, nature reserves and biodiversity sensitive areas, Production demand: Based on the existing industrial layout and resource utilization efficiency requirements, define the production suitability area. Living needs: Define living suitability areas through population density and public service coverage indicators.

[0013] As a preferred solution of the national land space planning and refined management method based on big data described in the present invention, in which: in step S3, the way of dividing the three living spaces is: The hierarchical AHP analysis method was used to construct a comprehensive evaluation model of suitability, calculate the suitability scores of ecology, production and life, and generate a spatial division map of the three ecosystems based on the grid distribution of the suitability scores.

[0014] As a preferred solution of the land space planning and refined management method based on big data described in the present invention, the steps of dividing the three living spaces by combining the resource classification map and the demand benchmark are as follows: Based on step S2 Calculate the resource suitability score using the following formula: ,in, For Region The comprehensive suitability score of For Region The ecological suitability score of For Region The production suitability score of For Region The suitability score of life are the weight factors of ecological, production and living needs, respectively, calculated by the analytic hierarchy process (AHP); Calculate a suitability score based on a need benchmark, including: The ecological suitability score is calculated as follows: , in, For Region of ecological protection area, is the ecological region priority coefficient, ranging from 0 to 1, with 1 representing the highest priority. For Region The total area, Production suitability score, calculated as: , in, For Region The total output value of existing industries, For Region The total resource consumption, The living suitability score is calculated as follows: , in, For Region population density, For Region The coverage rate of public service facilities; The analytic hierarchy process (AHP) was used to calculate the weight factors, and a pairwise comparison matrix of ecological, production and living needs was constructed according to the regional goals, which was set as: , Pair Matrix Perform normalization processing, the normalization formula is: , The weight factor is obtained by normalizing the eigenvector of the matrix. The calculation formula is: , in, is a weight factor vector containing ; Based on the comprehensive evaluation score of suitability , for the region To rasterize: Ecological space division conditions: , Production space division conditions: , Living space division conditions: , in, is the mean value of ecological space suitability score, is the mean of the production space suitability scores.

[0015] As a preferred solution of the national land space planning and refined management method based on big data described in the present invention, the step of using the linear programming LP model to perform multi-scenario simulation and output a division scheme that meets local goals is as follows: Define the optimization goal, set it according to the needs of scenario simulation in the form of linear programming, and the objective function is: , in, To optimize the target value, it represents the comprehensive suitability score of the three living spaces. The priority weights for ecology, production and living space, Eco-first scenario: , Production Priority Scenario: , Balanced development scenario: , It is a collection of ecological, production and living areas. For Region The corresponding ecological, production and living space areas; Define constraints, including, Total space area constraint: ,in, is the total area, Space ratio constraints: , , , in, The minimum ratio of ecological, production and living space. The highest proportion of space, Spatial connectivity constraints: , in, For space In the area The minimum area, is the minimum area threshold for connectivity, Spatial function conflict constraints: , in, For Region total area.

[0016] As a preferred solution of the land space planning and refined management method based on big data described in the present invention, the steps of capturing monitoring data including land use and environmental changes through remote sensing images and IoT devices, evaluating the difference between the monitoring data and the division scheme generated in step S4, updating the deviation result to the data set of step S1, and optimizing the weight factor and the parameters of the comprehensive suitability evaluation model are as follows: Remote sensing images are used to obtain land use changes. Assuming that the land use monitoring data for each area is , , in, For Region In time The land use type area set, For Region In time of forest area, For Region In time of cultivated land area, For Region In time of building land area; Use IoT devices to obtain environmental change data. Suppose the environmental monitoring data is , , in, For Region In time A collection of environmental change data, For Region In time The air quality index, For Region In time the amount of water resources available; Combine land use and environmental data to build a comprehensive monitoring dataset: ,in, For Region In time Comprehensive monitoring data of Compare the monitoring data with the three-life space division scheme generated in step S4, and calculate the deviation of each space. The calculation formula is: , in, For Region Type of space The area deviation, is the area in step S4 The planned area, For monitoring data area The actual area of According to the severity of the deviation, different regions are given weights, and the weight update formula is: , in, For Region The bias weight of Update the deviation result to the resource status data set and adjust the input data. The adjustment formula is: , in, For updated land use data, is the land use data before updating, Update the trend model of environmental change data. The update formula is: , in, For updated environmental change data, This is the environmental change data before the update. To monitor the deviation of environmental change data from planned environmental targets. Use monitoring data to update the input of the random forest model and dynamically adjust the weight factor. The random forest input is , ,in, For Region Actual socio-economic data; The output new weight factor is: ,in, is the updated weight factor, Based on the new data, the demand benchmark of the suitability evaluation model is adjusted, and the evaluation formula is: , in, is the adjusted weight factor.

[0017] The beneficial effects of the present invention are as follows: first, the present invention dynamically obtains the distribution changes of forest land, cultivated land and construction land through remote sensing images, and combines IoT devices to monitor ecological data such as air quality and water resources, and uses mobile signaling data to update population density and economic activity changes. Secondly, the random forest model is used to dynamically adjust the weight factor, and combined with factors such as land availability, ecological sensitivity and infrastructure carrying capacity, the resource carrying capacity is comprehensively quantified and divided into high, medium and low carrying capacity areas; the problems of fixed factor weights and fuzzy regional division in traditional evaluation are overcome, and then, combined with ecological protection, industrial layout and public service coverage benchmarks, the hierarchical analysis method AHP is used to calculate the suitability scores of ecology, production and life, and the precise division of space is achieved through the rasterization method to ensure that the planning results are highly consistent with actual needs; finally, the guided linear programming and dynamic feedback mechanism are used to enhance the executability and adaptability of the plan, optimize spatial conflicts through scenario simulation, and use remote sensing and IoT monitoring data to continuously adjust model parameters to achieve dynamic synchronization between planning and reality, effectively solving the problems of strong staticity and lagging adjustment in traditional planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0019] Figure 1 This is a flow chart of the land space planning and refined management method in Example 1. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Example 1, reference Figure 1 This embodiment provides a land space planning and refined management method based on big data, including the following steps: Step S1, acquiring planning data in real time, including land resources, ecological environment and socio-economic data, and constructing a resource status data set based on the planning data; In step S1, the land resources, ecological environment and socio-economic data are obtained in the following manner: Obtain land resource data through remote sensing image platforms, including forest land, cultivated land and building land cover types, Obtain real-time ecological and environmental data through meteorological monitoring stations and environmental monitoring IoT devices, including water resource distribution, air quality, and climate indicators. Obtain socio-economic data through statistical yearbooks and mobile signaling data, including population density, economic activity intensity, transportation network layout, and public facilities distribution; Step S2, based on the resource status data set, construct a multi-objective weight model to evaluate resource carrying capacity, and generate a regional resource carrying capacity classification map, in which high, medium and low carrying capacity areas are divided; The weight factors of the multi-objective weight model are dynamically adjusted based on historical planning performance data using a random forest regression model; In step S2, the multi-objective weight model integrates three dimensions of resource carrying capacity, including: Land resource carrying factor: land availability, Ecological environment carrying factors: environmental sensitivity, Socioeconomic carrying factors: population pressure and infrastructure; Based on the resource status data set, the steps of constructing a multi-objective weight model to evaluate resource carrying capacity and generate a regional resource carrying capacity classification map are as follows: The weighted linear model is used to construct the resource carrying capacity score, and the calculation formula is: , in, For Region The comprehensive score of resource carrying capacity is For Region The land resource carrying factor represents the proportion of available land. For Region The ecological environment carrying factor indicates ecological sensitivity. For Region The socioeconomic carrying factor represents population pressure and infrastructure level. are the weight factors of the three dimensions of land resources, ecological environment and social economy. Each factor is calculated based on the resource status data, and the calculation formula is: , in, For Region of available land area, For Region of total land area, , is the ecological sensitivity coefficient, ranging from 0 to 1, with 1 indicating extremely high sensitivity. Ecologically sensitive area The area of is the number of ecologically sensitive areas, , in, For Region The total population of For Region The carrying capacity of infrastructure; The threshold classification method The high, medium and low bearing capacity areas are divided as follows: High load-bearing capacity , The medium bearing capacity is , Low load capacity , in, For all regions The average value of For all regions The standard deviation of the generated classification map represents the distribution of regional carrying capacity in the form of a spatial grid; In step S2, the weight factor is dynamically adjusted as follows: Assume that there is a historical plan Regional samples, each sample records the actual planning results and the corresponding resource factor , define the model training objective as: ,in, Planning areas for history The actual planning results is the random forest regression model, is the random error term, The importance of each factor is calculated using the random forest model. , Factor The weight factor update formula for the relative importance of planning results is: , in, Factor Dynamic weight, substitute the dynamically adjusted weight factor into the formula In the process, the resource carrying capacity score is recalculated and the classification diagram is updated; Specifically, the weight factors are adaptively adjusted through the random forest model, and the dynamic response capability of the model is improved based on historical data. The definition of each load-bearing factor is refined, and the three-dimensional resource factors are integrated through a linear weighted model. The grading results are intuitive and easy to use while maintaining flexibility.

[0024] Step S3, combining the resource classification map and demand benchmark, dividing the three life spaces, including ecology, production and life; In step S3, the demand benchmark includes: Ecological needs: define ecological suitability areas based on ecological red lines, nature reserves and biodiversity sensitive areas, Production demand: Based on the existing industrial layout and resource utilization efficiency requirements, define the production suitability area. Living needs: Define living suitability areas through population density and public service coverage indicators; In step S3, the way of dividing the three-life space is as follows: A hierarchical AHP analysis method was used to construct a comprehensive suitability evaluation model, calculate the suitability scores of ecology, production and life, and generate a spatial division map of the three ecosystems based on the grid distribution of the suitability scores. Combining the resource classification diagram and demand benchmark, the steps to divide the three life spaces are as follows: Based on step S2 Calculate the resource suitability score using the following formula: ,in, For Region The comprehensive suitability score of For Region The ecological suitability score of For Region The production suitability score of For Region The suitability score of life are the weight factors of ecological, production and living needs, respectively, calculated by the analytic hierarchy process (AHP); Calculate a suitability score based on a need benchmark, including: The ecological suitability score is calculated as follows: , in, For Region of ecological protection area, is the ecological region priority coefficient, ranging from 0 to 1, with 1 representing the highest priority. For Region The total area, Production suitability score, calculated as: , in, For Region The total output value of existing industries, For Region The total resource consumption, The living suitability score is calculated as follows: , in, For Region population density, For Region The coverage rate of public service facilities; The analytic hierarchy process (AHP) was used to calculate the weight factors, and a pairwise comparison matrix of ecological, production and living needs was constructed according to the regional goals, which was set as: , Pair Matrix Perform normalization processing, the normalization formula is: , The weight factor is obtained by normalizing the eigenvector of the matrix. The calculation formula is: , in, is a weight factor vector containing ; Based on the comprehensive evaluation score of suitability , for the region To rasterize: Ecological space division conditions: , Production space division conditions: , Living space division conditions: , in, is the mean value of ecological space suitability score, is the mean of the production space suitability scores; Specifically, through the benchmarks of ecological protection, production efficiency and life service coverage, the suitability calculation of the three life spaces is refined, and the AHP method is introduced to dynamically adjust the weights according to regional goals to adapt to different needs. The spatial division is achieved through score distribution, which is intuitive and easy to operate.

[0025] Step S4, using a linear programming LP model to perform multi-scenario simulation and output a partitioning scheme that meets local goals; The steps of using the linear programming LP model to perform multi-scenario simulation and output a partitioning scheme that meets local goals are as follows: Define the optimization goal, set it according to the needs of scenario simulation in the form of linear programming, and the objective function is: , in, To optimize the target value, it represents the comprehensive suitability score of the three living spaces. The priority weights for ecology, production and living space, Eco-first scenario: , Production Priority Scenario: , Balanced development scenario: , It is a collection of ecological, production and living areas. For Region The corresponding ecological, production and living space areas; Define constraints, including, Total space area constraint: ,in, is the total area, Space ratio constraints: , , , in, The minimum ratio of ecological, production and living space. The highest proportion of space, Spatial connectivity constraints: , in, For space In the area The minimum area, is the minimum area threshold for connectivity, Spatial function conflict constraints: , in, For Region Total area; For different goals, set priority weights and proportion constraints to simulate multiple scenarios, including: Eco-priority scenario, Prioritize the protection of ecological space: , Limit the production space ratio: ; Production priority scenario, Prioritize the development of production space: , Limit the proportion of ecological space: ; Balanced development scenario, Ecological, production and living space are given equal weights: , The proportions of each space are close to balanced: ; Obtain the optimal solution through linear programming , rasterize the optimization results to generate the final three-life space division scheme; Specifically, by adjusting priority weights and proportional constraints, it can adapt to multiple objectives, optimize the spatial distribution of the three lives based on regional needs and resource characteristics, use linear programming tools to quickly solve, achieve dynamic integration of guidance and spatial optimization, and effectively guide national land space planning in complex scenarios.

[0026] Step S5, capturing monitoring data, including land use and environmental changes, through remote sensing images and IoT devices, evaluating the difference between the monitoring data and the partitioning scheme generated in step S4, updating the deviation results to the data set of step S1, and optimizing the weight factors and parameters of the comprehensive suitability evaluation model; The monitoring data, including land use and environmental changes, are captured through remote sensing images and IoT devices, the difference between the monitoring data and the partitioning scheme generated in step S4 is evaluated, and the deviation results are updated to the data set of step S1. The steps of optimizing the weight factors and the parameters of the comprehensive suitability evaluation model are as follows: Remote sensing images are used to obtain land use changes. Assuming that the land use monitoring data for each area is , , in, For Region In time The land use type area set, For Region In time of forest area, For Region In time of cultivated land area, For Region In time of building land area; Use IoT devices to obtain environmental change data. Suppose the environmental monitoring data is , , in, For Region In time A collection of environmental change data, For Region In time The air quality index, For Region In time the amount of water resources available; Combine land use and environmental data to build a comprehensive monitoring dataset: ,in, For Region In time Comprehensive monitoring data of Compare the monitoring data with the three-life space division scheme generated in step S4, and calculate the deviation of each space. The calculation formula is: , in, For Region Type of space The area deviation, is the area in step S4 The planned area, For monitoring data area The actual area of According to the severity of the deviation, different regions are given weights, and the weight update formula is: , in, For Region The bias weight of Update the deviation result to the resource status data set and adjust the input data. The adjustment formula is: , in, For updated land use data, is the land use data before updating, Update the trend model of environmental change data. The update formula is: , in, For updated environmental change data, This is the environmental change data before the update. To monitor the deviation of environmental change data from planned environmental targets. Use monitoring data to update the input of the random forest model and dynamically adjust the weight factor. The random forest input is , ,in, For Region Actual socio-economic data; The output new weight factor is: ,in, is the updated weight factor, Based on the new data, the demand benchmark of the suitability evaluation model is adjusted, and the evaluation formula is: , in, is the adjusted weight factor; Specifically, remote sensing and IoT data are used to dynamically capture changes and refine spatial management. Deviation weights are used to adjust weight factors and suitability model parameters to enhance the dynamic response capability of planning, realize a closed loop of data collection, deviation assessment and model optimization, and effectively combine planning with dynamic monitoring.

[0027] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for land space planning and refined management based on big data, characterized by: include, Step S1, acquiring planning data in real time, including land resources, ecological environment and socio-economic data, and constructing a resource status data set based on the planning data; Step S2, based on the resource status data set, construct a multi-objective weight model to evaluate resource carrying capacity, and generate a regional resource carrying capacity classification map, in which high, medium and low carrying capacity areas are divided; Step S3, combining the resource classification map and demand benchmark, dividing the three life spaces, including ecology, production and life; Step S4, using a linear programming LP model to perform multi-scenario simulation and output a partitioning scheme that meets local goals; In step S5, monitoring data including land use and environmental changes are captured through remote sensing images and IoT devices, the difference between the monitoring data and the partitioning scheme generated in step S4 is evaluated, the deviation results are updated to the data set of step S1, and the weight factors and parameters of the comprehensive suitability evaluation model are optimized.

2. The method for land space planning and refined management based on big data according to claim 1, characterized in that: In step S1, the land resources, ecological environment and socio-economic data are obtained in the following manner: Obtain land resource data through remote sensing image platforms, including forest land, cultivated land and building land cover types, Obtain real-time ecological and environmental data through meteorological monitoring stations and environmental monitoring IoT devices, including water resource distribution, air quality, and climate indicators. Socioeconomic data, including population density, intensity of economic activities, layout of transportation networks, and distribution of public facilities, are obtained through statistical yearbooks and mobile signaling data.

3. The method for land space planning and refined management based on big data as claimed in claim 2, characterized in that: The weight factors of the multi-objective weight model are dynamically adjusted according to historical planning performance data using a random forest regression model.

4. The method for land space planning and refined management based on big data as claimed in claim 3, characterized in that: In step S2, the multi-objective weight model integrates three dimensions of resource carrying capacity, including: Land resource carrying factor: land availability, Ecological environment carrying factors: environmental sensitivity, Socioeconomic carrying factors: population pressure and infrastructure.

5. The method for land space planning and refined management based on big data according to claim 4, characterized in that: The steps of constructing a multi-objective weight model to evaluate resource carrying capacity based on the resource status data set and generating a regional resource carrying capacity classification map are as follows: The weighted linear model is used to construct the resource carrying capacity score, and the calculation formula is: ， in, For Region The comprehensive score of resource carrying capacity is For Region The land resource carrying factor represents the proportion of available land. For Region The ecological environment carrying factor indicates ecological sensitivity. For Region The socioeconomic carrying factor represents population pressure and infrastructure level. are the weight factors of the three dimensions of land resources, ecological environment and social economy. Each factor is calculated based on the resource status data, and the calculation formula is: ， in, For Region of available land area, For Region of total land area, ， is the ecological sensitivity coefficient, ranging from 0 to 1, with 1 indicating extremely high sensitivity. Ecologically sensitive area The area of is the number of ecologically sensitive areas, ， in, For Region The total population of For Region The carrying capacity of infrastructure; The threshold classification method The high, medium and low bearing capacity areas are divided as follows: High load-bearing capacity , The medium bearing capacity is , Low load capacity , in, For all regions The average value of For all regions The standard deviation of the generated classification map represents the distribution of regional carrying capacity in the form of a spatial grid; In step S2, the weight factor is dynamically adjusted as follows: Assume that there is a historical plan Regional samples, each sample records the actual planning results and the corresponding resource factor , define the model training objective as: ,in, Planning areas for history The actual planning results is the random forest regression model, is the random error term, The importance of each factor is calculated using the random forest model. , Factor The weight factor update formula for the relative importance of planning results is: ， in, Factor Dynamic weight, substitute the dynamically adjusted weight factor into the formula In the process, the resource carrying capacity score is recalculated and the classification diagram is updated.

6. The method for land space planning and refined management based on big data according to claim 5, characterized in that: In step S3, the demand benchmark includes: Ecological needs: define ecological suitability areas based on ecological red lines, nature reserves and biodiversity sensitive areas, Production demand: Based on the existing industrial layout and resource utilization efficiency requirements, define the production suitability area. Living needs: Define living suitability areas through population density and public service coverage indicators.

7. The method for land space planning and refined management based on big data according to claim 6, characterized in that: In step S3, the way of dividing the three-life space is as follows: The hierarchical AHP analysis method was used to construct a comprehensive evaluation model of suitability, calculate the suitability scores of ecology, production and life, and generate a spatial division map of the three ecosystems based on the grid distribution of the suitability scores.

8. The method for land space planning and refined management based on big data according to claim 7, characterized in that: The steps of dividing the three life spaces by combining the resource classification diagram and the demand benchmark are as follows: Based on step S2 Calculate the resource suitability score using the following formula: ,in, For Region The comprehensive suitability score of For Region The ecological suitability score of For Region The production suitability score of For Region The suitability score of life are the weight factors of ecological, production and living needs, respectively, calculated by the analytic hierarchy process (AHP); Calculate a suitability score based on a need benchmark, including: The ecological suitability score is calculated as follows: ， in, For Region of ecological protection area, is the ecological region priority coefficient, ranging from 0 to 1, with 1 representing the highest priority. For Region The total area, Production suitability score, calculated as: ， in, For Region The total output value of existing industries, For Region The total resource consumption, The living suitability score is calculated as follows: ， in, For Region population density, For Region The coverage rate of public service facilities; The analytic hierarchy process (AHP) was used to calculate the weight factors, and a pairwise comparison matrix of ecological, production and living needs was constructed according to the regional goals, which was set as: ， Pair Matrix Perform normalization processing, the normalization formula is: ， The weight factor is obtained by normalizing the eigenvector of the matrix. The calculation formula is: ， in, is a weight factor vector containing ; Based on the comprehensive evaluation score of suitability , for the region To rasterize: Ecological space division conditions: , Production space division conditions: , Living space division conditions: , in, is the mean value of ecological space suitability score, is the mean of the production space suitability scores.

9. The method for land space planning and refined management based on big data according to claim 8, characterized in that: The steps of using the linear programming LP model to perform multi-scenario simulation and output a partitioning scheme that meets local goals are as follows: Define the optimization goal, set it according to the needs of scenario simulation in the form of linear programming, and the objective function is: ， in, To optimize the target value, it represents the comprehensive suitability score of the three living spaces. The priority weights for ecology, production and living space, Eco-first scenario: , Production Priority Scenario: , Balanced development scenario: , It is a collection of ecological, production and living areas. For Region The corresponding ecological, production and living space areas; Define constraints, including, Total space area constraints: ,in, is the total area, Space ratio constraints: ， ， ， in, The minimum ratio of ecological, production and living space. The highest proportion of space, Spatial connectivity constraints: ， in, For space In the area The minimum area, is the minimum area threshold for connectivity, Spatial function conflict constraints: ， in, For Region total area.

10. The method for land space planning and refined management based on big data according to claim 9, characterized in that: The steps of capturing monitoring data including land use and environmental changes through remote sensing images and IoT devices, evaluating the difference between the monitoring data and the partitioning scheme generated in step S4, updating the deviation results to the data set of step S1, and optimizing the weight factors and parameters of the comprehensive suitability evaluation model are as follows: Remote sensing images are used to obtain land use changes. Assuming that the land use monitoring data for each area is , ， in, For Region In time The land use type area set, For Region In time of forest area, For Region In time of cultivated land area, For Region In time of building land area; Use IoT devices to obtain environmental change data. Suppose the environmental monitoring data is , ， in, For Region In time A collection of environmental change data, For Region In time The air quality index, For Region In time the amount of water resources available; Combine land use and environmental data to build a comprehensive monitoring dataset: ,in, For Region In time Comprehensive monitoring data of Compare the monitoring data with the three-life space division scheme generated in step S4, and calculate the deviation of each space. The calculation formula is: ， in, For Region Type of space The area deviation, is the area in step S4 The planned area, For monitoring data area The actual area of According to the severity of the deviation, different regions are given weights, and the weight update formula is: ， in, For Region The bias weight of Update the deviation result to the resource status data set and adjust the input data. The adjustment formula is: ， in, For updated land use data, is the land use data before updating, Update the trend model of environmental change data. The update formula is: ， in, is the updated environmental change data, This is the environmental change data before the update. To monitor the deviation of environmental change data from planned environmental targets. Use monitoring data to update the input of the random forest model and dynamically adjust the weight factor. The random forest input is , ,in, For Region Actual socio-economic data; The output new weight factor is: ,in, is the updated weight factor, Based on the new data, the demand benchmark of the suitability evaluation model is adjusted, and the evaluation formula is: ， in, is the adjusted weight factor.

Images