Construction land scale prediction method based on territorial space planning

Abstract

The invention relates to the technical field of territorial space planning, in particular to a construction land scale prediction method based on territorial space planning. The method comprises the following steps: acquiring historical construction land data, regional economic index data and control line space data of different regions; performing potential development turning point identification according to historical construction land data and regional economic index data, and performing stage trend analysis to obtain dynamic stage trend data; multi-scene prediction is carried out according to the dynamic stage trend data, so that multi-scene prediction data are obtained, and multiple scenes comprise a reference scene, a transformation scene and an innovation scene; and constructing an industry-land demand sensitivity analysis model according to the regional economic index data and the historical construction land data. According to the method, through development turning point identification and stage trend analysis, the dynamic characteristics of construction land change can be better grasped.

Description

Technical Field

[0001] The present invention relates to the technical field of national land space planning, and in particular to a method for predicting the scale of construction land based on national land space planning. Background Art

[0002] National land space planning is the overall arrangement and layout of the development, protection, utilization and improvement of the national land space of a country or region. Its goal is to optimize the national land space layout and improve the efficiency of land use. Its main contents include spatial layout planning (such as urban system planning, industrial layout planning, infrastructure layout planning and ecological space protection planning), land classification and control (such as agricultural land, construction land, unused land classification and various types of land control measures and indicators), ecological environment protection and resource utilization and protection. The scale of construction land is an important concept in national land space planning. It specifically refers to the total amount of land used for various construction activities in a certain area (such as a city, county or province). It is also the total area of ​​land allowed to be used for urban and rural construction and various development activities during the planning period.

[0003] Traditional methods for predicting the scale of construction land based on national land space planning often have the following problems: too much historical inertia, over-reliance on historical trends, and neglect of possible turning points in the future; based on the past high-speed expansion trend, it failed to foresee in time the trend of urban development shifting from external expansion to internal quality improvement; it ignored the changes in land demand brought about by industrial structure adjustment and still predicted according to the traditional industrial land use model. It is difficult to accurately quantify the impact of the "three control lines" (ecological protection red line, permanent basic farmland, and urban development boundary) on the scale of construction land. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a method for predicting the scale of construction land based on national land space planning to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above purpose, a construction land scale prediction method based on national land space planning includes the following steps:

[0006] Step S1: Obtain historical construction land data, regional economic indicator data, and control line spatial data of different regions; identify potential development turning points based on historical construction land data and regional economic indicator data, and perform stage trend analysis to obtain dynamic stage trend data;

[0007] Step S2: Perform multi-scenario forecasting based on dynamic stage trend data to obtain multi-scenario forecasting data, where the multi-scenario includes a baseline scenario, a transformation scenario, and an innovation scenario; construct an industry-land demand sensitivity analysis model based on regional economic indicator data and historical construction land data; generate possible land demand forecast ranges for different industries based on the industry-land demand sensitivity analysis model to obtain land demand elasticity forecasting data;

[0008] Step S3: Use GIS technology to perform spatial analysis on the available land under the constraints of the control line spatial data, so as to obtain the data of the buildable area; use a multi-objective optimization algorithm to optimize the land use layout according to the buildable area data and multi-scenario prediction data, so as to obtain the optimized layout plan data;

[0009] Step S4: integrating the multi-scenario prediction data, the land demand elasticity prediction data and the optimized layout scheme data based on the Bayesian network model, and performing uncertainty assessment to obtain elasticity prediction result data;

[0010] Step S5: Acquire planning indicator data of different planning scales; construct a multi-scale land demand coordination model based on the elastic prediction result data and the planning indicator data, so as to obtain coordinated land scale prediction data.

[0011] By collecting historical construction land data and regional economic indicators, the embodiment of the present invention can identify turning points and trends in regional development and provide basic information for subsequent planning; stage trend analysis helps understand the dynamic process of regional development and provides a basis for formulating reasonable development strategies. Multi-scenario forecasting can provide development possibilities under different assumptions and help decision makers examine issues from multiple perspectives; the industry-land demand sensitivity analysis model can evaluate the impact of different industrial development on land demand and provide a scientific basis for the rational allocation of land resources. Using GIS technology for spatial analysis can more intuitively identify available land resources and ensure the feasibility of planning; the application of multi-objective optimization algorithms can balance multiple development goals, such as economic benefits, environmental impacts, etc., to achieve the optimal configuration of land use layout. Integrating data from different sources through the Bayesian network model can improve the accuracy and reliability of predictions; uncertainty assessment helps identify potential risks in planning and provide a more comprehensive perspective for decision-making. Constructing a multi-scale land demand coordination model can ensure consistency and coordination at different planning levels; this helps achieve sustainable development of the region as a whole while taking into account local special needs and conditions. These steps together constitute a systematic and scientific planning process, which aims to achieve the following goals: reasonably allocate land resources and improve land use efficiency; reduce risks in the planning implementation process through uncertainty assessment; ensure that the planning scheme meets long-term economic development and environmental protection goals; provide decision makers with multi-angle and multi-scenario analysis results to assist in making more scientific and reasonable decisions. The entire process emphasizes data-driven decision-making, and achieves regional balance and sustainable development by comprehensively considering economic, social, environmental and other factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0013] Figure 1 It is a schematic diagram of the steps of the method for predicting the scale of construction land based on national land space planning of the present invention;

[0014] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0015] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0016] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0017] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0018] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0019] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for predicting the scale of construction land based on national land space planning, the method comprising the following steps:

[0020] Step S1: Obtain historical construction land data, regional economic indicator data, and control line spatial data of different regions; identify potential development turning points based on historical construction land data and regional economic indicator data, and perform stage trend analysis to obtain dynamic stage trend data;

[0021] Step S2: Perform multi-scenario forecasting based on dynamic stage trend data to obtain multi-scenario forecasting data, where the multi-scenario includes a baseline scenario, a transformation scenario, and an innovation scenario; construct an industry-land demand sensitivity analysis model based on regional economic indicator data and historical construction land data; generate possible land demand forecast ranges for different industries based on the industry-land demand sensitivity analysis model to obtain land demand elasticity forecasting data;

[0022] Step S3: Use GIS technology to perform spatial analysis on the available land under the constraints of the control line spatial data, so as to obtain the data of the buildable area; use a multi-objective optimization algorithm to optimize the land use layout according to the buildable area data and multi-scenario prediction data, so as to obtain the optimized layout plan data;

[0023] Step S4: integrating the multi-scenario prediction data, the land demand elasticity prediction data and the optimized layout scheme data based on the Bayesian network model, and performing uncertainty assessment to obtain elasticity prediction result data;

[0024] Step S5: Acquire planning indicator data of different planning scales; construct a multi-scale land demand coordination model based on the elastic prediction result data and the planning indicator data, so as to obtain coordinated land scale prediction data.

[0025] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a method for predicting the scale of construction land based on national land space planning of the present invention. In this example, the method for predicting the scale of construction land based on national land space planning includes the following steps:

[0026] Step S1: Obtain historical construction land data, regional economic indicator data, and control line spatial data of different regions; identify potential development turning points based on historical construction land data and regional economic indicator data, and perform stage trend analysis to obtain dynamic stage trend data;

[0027] The embodiment of the present invention collects and organizes the data on the use of construction land in a region within a certain time range in the past, including information such as land use type, area, location, and development intensity; collects data related to economic development in the region, such as GDP, industrial structure, population growth, investment, etc., which can reflect the trend of economic development and its impact on land demand; obtains geospatial data involving ecological protection red lines, permanent basic farmland, and urban development boundaries (i.e., "three control lines"). These data are used to determine the scope of developable land. Using time series analysis and machine learning methods, historical construction land and regional economic data are analyzed to identify key turning points that may exist in past development. These turning points may mark changes in economic structure, policy adjustments, or other major events that affect land use. After identifying the turning points, the historical data is divided into different stages, and the land use trend of each stage is analyzed; the analysis content includes the speed of land development, changes in land use, and the impact of economic indicators on land demand, thereby generating dynamic stage trend data.

[0028] Step S2: Perform multi-scenario forecasting based on dynamic stage trend data to obtain multi-scenario forecasting data, where the multi-scenario includes a baseline scenario, a transformation scenario, and an innovation scenario; construct an industry-land demand sensitivity analysis model based on regional economic indicator data and historical construction land data; generate possible land demand forecast ranges for different industries based on the industry-land demand sensitivity analysis model to obtain land demand elasticity forecasting data;

[0029] The embodiment of the present invention assumes that future development follows historical trends and constructs baseline scenario forecast data. The baseline scenario is usually based on the continuation of current policies and economic conditions; considering possible economic structural adjustments, industrial upgrades or technological progress, a transformation scenario is constructed. This scenario focuses on the development path after the turning point and the impact on land demand; assuming that there will be innovative changes in future development, such as new economic models and the rise of new industries, innovative scenario forecast data is constructed. This scenario assumption is usually based on the positive impact of technological innovation and institutional changes. Based on regional economic indicator data and historical construction land data, a sensitivity analysis model reflecting the relationship between industrial development and land demand is constructed. The model can quantify the elasticity of land demand for different industries, that is, the corresponding changes in land demand when economic conditions change. Using the sensitivity analysis model, the industrial development under different scenarios is simulated, the possible land demand range of each industry is predicted, and the land demand elasticity prediction data is obtained.

[0030] Step S3: Use GIS technology to perform spatial analysis on the available land under the constraints of the control line spatial data, so as to obtain the data of the buildable area; use a multi-objective optimization algorithm to optimize the land use layout according to the buildable area data and multi-scenario prediction data, so as to obtain the optimized layout plan data;

[0031] The embodiment of the present invention uses a geographic information system (GIS) to analyze the acquired control line spatial data to identify the available land subject to the "three control lines". This analysis will clarify the land area that can be used for future development; through spatial analysis, the restricted area is distinguished from the developable area, and the buildable area data is generated. On the basis of the buildable area data, combined with multi-scenario prediction data, a multi-objective optimization algorithm is used to optimize the land layout. The optimization goals may include maximizing land use efficiency, minimizing the impact on the ecological environment, and improving traffic convenience. Based on the results of the optimization algorithm, land layout plan data under different scenarios are generated to ensure efficient and sustainable use of land resources.

[0032] Step S4: integrating the multi-scenario prediction data, the land demand elasticity prediction data and the optimized layout scheme data based on the Bayesian network model, and performing uncertainty assessment to obtain elasticity prediction result data;

[0033] The embodiment of the present invention uses a Bayesian network model to integrate multi-scenario prediction data, land demand elasticity prediction data, and optimized layout plan data. The Bayesian network can handle complex probability relationships, integrate data from different sources, and generate consistent prediction results. Through the reasoning ability of the Bayesian network, the uncertainty in the prediction results is evaluated. This step identifies possible deviations and risks in the prediction process and ensures that the final prediction results are reliable and robust. Based on the results of the uncertainty assessment, elastic prediction result data is generated to provide a multi-dimensional reference basis for decision-making.

[0034] Step S5: Acquire planning indicator data of different planning scales; construct a multi-scale land demand coordination model based on the elastic prediction result data and the planning indicator data, so as to obtain coordinated land scale prediction data.

[0035] The embodiment of the present invention obtains planning indicator data at different scales, such as regional planning, urban planning, and county-level planning. These indicators include land use intensity, ecological protection requirements, population distribution, etc. Based on the elastic prediction result data and planning indicator data, a multi-scale land demand coordination model is constructed. The model aims to coordinate the land demand between different levels of planning to ensure planning consistency and rationality. Through the operation of the model, coordinated land scale prediction data is generated. These data can guide future land use planning at different scales and achieve rational allocation of resources and sustainable development.

[0036] By collecting historical construction land data and regional economic indicators, the embodiment of the present invention can identify turning points and trends in regional development and provide basic information for subsequent planning; stage trend analysis helps understand the dynamic process of regional development and provides a basis for formulating reasonable development strategies. Multi-scenario forecasting can provide development possibilities under different assumptions and help decision makers examine issues from multiple perspectives; the industry-land demand sensitivity analysis model can evaluate the impact of different industrial development on land demand and provide a scientific basis for the rational allocation of land resources. Using GIS technology for spatial analysis can more intuitively identify available land resources and ensure the feasibility of planning; the application of multi-objective optimization algorithms can balance multiple development goals, such as economic benefits, environmental impacts, etc., to achieve the optimal configuration of land use layout. Integrating data from different sources through the Bayesian network model can improve the accuracy and reliability of predictions; uncertainty assessment helps identify potential risks in planning and provide a more comprehensive perspective for decision-making. Constructing a multi-scale land demand coordination model can ensure consistency and coordination at different planning levels; this helps achieve sustainable development of the region as a whole while taking into account local special needs and conditions. These steps together constitute a systematic and scientific planning process, which aims to achieve the following goals: reasonably allocate land resources and improve land use efficiency; reduce risks in the planning implementation process through uncertainty assessment; ensure that the planning scheme meets long-term economic development and environmental protection goals; provide decision makers with multi-angle and multi-scenario analysis results to assist in making more scientific and reasonable decisions. The entire process emphasizes data-driven decision-making, and achieves regional balance and sustainable development by comprehensively considering economic, social, environmental and other factors.

[0037] Preferably, step S1 comprises the following steps:

[0038] Step S11: Obtain historical construction land data, including the area, distribution and changes of different types of construction land;

[0039] Step S12: Collecting regional GDP, fixed asset investment, industrial structure ratio, population size and composition, and aligning them with historical construction land data to obtain regional economic indicator data;

[0040] Step S13: Acquire spatial data of control lines in different regions, including GIS layer data of ecological protection red lines, permanent basic farmland lines, and urban development boundaries;

[0041] Step S14: cleaning and standardizing the historical construction land data and regional economic indicator data to remove outliers and missing values, thereby obtaining standard construction land data;

[0042] Step S15: using time series decomposition technology to separate the trend, seasonality and random components of the construction land standard data, so as to obtain the basic trend data of the construction land change;

[0043] Step S16: Identify potential development turning points of the basic trend data and perform stage trend analysis to obtain dynamic stage trend data.

[0044] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps:

[0045] Step S11: Obtain historical construction land data, including the area, distribution and changes of different types of construction land;

[0046] The embodiments of the present invention collect different types of construction land data, including residential land, commercial land, industrial land, infrastructure land, public service land, etc.; count the area and geographical distribution of these land types at different time points, identify the change trend and expansion direction; record the changes in each type of land, including addition, reduction, change of use, etc. These data can come from the annual report of the land management department, cadastral data, remote sensing image analysis, etc.

[0047] Step S12: Collecting regional GDP, fixed asset investment, industrial structure ratio, population size and composition, and aligning them with historical construction land data to obtain regional economic indicator data;

[0048] The embodiment of the present invention obtains GDP data corresponding to the historical construction land data, usually in annual units, to reflect changes in the total regional economy.

[0049] Economic structure segmentation: If possible, obtain more detailed GDP data, decomposed into the output value of the primary, secondary and tertiary industries, to analyze the impact of economic structure on land demand. Collect fixed asset investment data for each year, especially investment data related to real estate development, infrastructure construction and industrial projects, which directly affect land demand. Count the proportion of each industry (such as industry, services, agriculture, etc.) in the regional economy, as well as the historical changes in these proportions, to understand the impact of industrial evolution trends on land use. Collect the total population and its composition in the region, including age structure, gender ratio, registered and non-registered population, etc. These data will help predict future residential and public service land demand. Align the above economic indicator data with historical construction land data, that is, ensure that the data is compared and analyzed at the same time node, so as to generate regional economic indicator data.

[0050] Step S13: Acquire spatial data of control lines in different regions, including GIS layer data of ecological protection red lines, permanent basic farmland lines, and urban development boundaries;

[0051] The embodiment of the present invention obtains GIS layer data of ecological protection red lines, which define ecological areas that need to be protected and restrict development activities; obtains GIS layer data of permanent basic farmland, which should be kept for agricultural use and cannot be converted into construction land; obtains GIS layer data of urban development boundaries, clarifies the spatial scope of urban expansion, and avoids disorderly urban expansion. The control line GIS layer data of different regions are integrated to ensure that various restricted areas can be clearly identified in subsequent analysis to guide the prediction of construction land.

[0052] Step S14: cleaning and standardizing the historical construction land data and regional economic indicator data to remove outliers and missing values, thereby obtaining standard construction land data;

[0053] The embodiment of the present invention analyzes historical construction land and regional economic indicator data, identifies and removes outliers in the data, such as land area that obviously exceeds a reasonable range or abnormal fluctuations in economic indicators. For missing data, a reasonable filling method is used, such as interpolation using data from adjacent time nodes, or using mean filling and other methods to ensure the integrity of the data. Different types of economic data are standardized to be in the same order of magnitude or unit to facilitate subsequent comprehensive analysis. This step can use normalization, Z-score standardization and other methods. Through data cleaning and standardization, clean, unified format construction land standard data is obtained, laying the foundation for subsequent analysis.

[0054] Step S15: using time series decomposition technology to separate the trend, seasonality and random components of the construction land standard data, so as to obtain the basic trend data of the construction land change;

[0055] The embodiment of the present invention uses time series decomposition technology to decompose the standard data of construction land into long-term trend components, revealing the long-term growth or decline trend of land use; separate seasonal components to capture regular fluctuations caused by seasonal changes, policy cycles, etc.; identify and separate random components in the data, which represent short-term, irregular fluctuations, which may be the impact of incidental events or temporary policies. Extract basic trend data from the decomposed data as the core trend line reflecting the changes in construction land, providing a basis for subsequent analysis and prediction.

[0056] Step S16: Identify potential development turning points of the basic trend data and perform stage trend analysis to obtain dynamic stage trend data.

[0057] The embodiment of the present invention uses statistical methods or machine learning algorithms such as Chow Test and CUSUM (cumulative sum) to identify potential turning points in basic trend data. These turning points may correspond to economic crises, major policy changes, technological advances or other important events, and mark these turning points on the timeline to provide a basis for stage trend analysis. According to the identified turning points, the historical data is divided into different stages, each of which may represent a different economic cycle or development stage. The construction land trend in each stage is independently modeled and analyzed to generate stage-by-stage dynamic trend data. These data can more accurately reflect the changes in land demand in each development stage and provide a more detailed basis for future scenario forecasting.

[0058] The present invention understands the area and distribution of different types of construction land, analyzes land use patterns and historical changes, evaluates land use efficiency, identifies possible inefficient or overdeveloped areas, and helps make evidence-based decisions. Through indicators such as regional GDP and fixed asset investment, the level of regional economic development is evaluated, and the relationship between population growth and economic development is analyzed in combination with population data. Based on economic and population data, future land demand is predicted to provide a basis for land supply planning. Ensure that the ecological protection red line is observed, protect the ecological environment and biodiversity, clarify the boundaries of urban development, prevent disorderly expansion, promote intensive land use, coordinate planning needs in different regions, and ensure the consistency and coordination of planning. By cleaning and removing erroneous and incomplete data, the reliability of data is improved, the data format is unified, and the comparison and analysis between different data sets is facilitated, avoiding the influence of outliers and missing values ​​on the analysis results. Separate long-term trends, identify the long-term development direction of land use, identify seasonal changes, understand the changing pattern of land use in a specific period, evaluate the impact of random factors on land use changes, and provide information for risk management. Identify key periods of land use change, such as policy changes, economic crises, etc., and divide land use changes into different stages according to turning points, so as to analyze the characteristics and driving factors of each stage. Based on stage trend analysis, predict future land use change trends and provide forward-looking guidance for planning. The implementation of these steps helps to build a comprehensive and detailed land use and economic development analysis framework. They not only provide an in-depth understanding of past and present land use, but also provide a scientific basis for predicting future trends and making corresponding decisions. Through these steps, planners can better understand the relationship between land use and economic development, and develop planning schemes that meet current needs while taking into account long-term sustainable development.

[0059] Preferably, step S16 comprises the following steps:

[0060] Step S161: using the Bayesian change point detection method to identify potential development turning points of basic trend data, thereby obtaining key turning point data;

[0061] The embodiment of the present invention selects a Bayesian change point detection method, such as Bayesian Online Change Point Detection, which is a statistical method that uses Bayesian theory to detect change points in time series data; based on basic trend data, a Bayesian change point detection model is constructed, which calculates the probability that each point in the time series is a turning point by estimating the prior and likelihood of the data. The model is run to analyze the basic trend data to identify potential development turning points, which may represent major changes in history, such as economic crises, policy adjustments, or industrial upgrades. The detection results are analyzed to screen out turning points with high probability to form key turning point data, which will be used for subsequent stage division and analysis.

[0062] Step S162: dividing the basic trend data into different development stages according to the key turning point data, and extracting features for each stage, thereby obtaining stage feature data;

[0063] The embodiment of the present invention uses the key turning point data obtained in step S161 to divide the basic trend data into several development stages, each stage represents a different cycle or economic stage of land use change; the start and end time of each stage are marked on the time series to form stage division data. Feature extraction is performed on each development stage, including the speed of land use change, the magnitude of change, the land use conversion rate, the economic growth rate, the industrial structure change, etc. The extracted features are summarized to form stage feature data. These data are used to describe the main characteristics and development trends of each stage.

[0064] Step S163: using a gradient boosting tree to perform a nonlinear relationship analysis between regional economic indicators and changes in construction land according to stage characteristic data, thereby obtaining indicator impact weight data;

[0065] The embodiment of the present invention selects Gradient Boosting Decision Trees (GBDT) as the main analysis tool because of its superior performance in processing nonlinear relationships and high-dimensional data. The stage characteristic data is combined with the regional economic indicator data as the input data set. The input data includes stage characteristics (such as land change rate, economic growth rate) and economic indicators (such as GDP, industry proportion, population growth, etc.). The data is trained using the gradient boosting tree model. The model is iteratively optimized and gradually weighted prediction errors to capture the complex nonlinear relationship between regional economic indicators and changes in construction land; after the model training is completed, the weights of the impact of each economic indicator on the change in construction land are extracted. The weight reflects the relative importance of different economic indicators to land demand.

[0066] Step S164: constructing a dynamic trend prediction model based on the Prophet model according to the stage characteristic data and the indicator influence weight data;

[0067] Prophet, an embodiment of the present invention, is a time series prediction model developed by Facebook. It is good at processing time series data with strong seasonality or periodicity and is suitable for dynamic trend prediction in various scenarios. The stage characteristic data is combined with the indicator influence weight data as the input data of the Prophet model. These data are used to build a prediction model to capture changes in future trends. According to the characteristics of historical data, the relevant parameters of the Prophet model are set, including seasonal components, trend components, and holiday components; the Prophet model is trained using the input data, and the model is tuned by cross-validation and other methods to ensure that it can accurately reflect historical trends and predict future changes.

[0068] Step S165: Use the dynamic trend prediction model to make a preliminary prediction of the trend of construction land changes in a certain period in the future, so as to obtain dynamic stage trend data.

[0069] The embodiment of the present invention uses a trained dynamic trend prediction model to predict the trend of construction land changes in a certain period of time in the future (such as 5 years or 10 years), including changes in land demand at different stages in the future, possible turning points, and seasonal fluctuations; the prediction results are organized into dynamic stage trend data, including possible future land use patterns, regional development trends, etc. The dynamic stage trend data is visualized, such as generating trend graphs, stage prediction charts, etc., to facilitate subsequent decision analysis and planning adjustments; preliminary analysis is performed based on the prediction results to identify possible risks and opportunities, and provide data support for subsequent detailed analysis and optimization.

[0070] The present invention can more accurately identify potential development turning points in historical construction land data through the Bayesian change point detection method. These key turning points are important bases for understanding land use change patterns and the motivations behind them. Dividing the basic trend data into different development stages and extracting the characteristics of each stage can help to deeply understand the characteristics and laws of land use changes in each stage, and provide support for the formulation of targeted planning strategies. The nonlinear relationship between regional economic indicators and construction land changes is analyzed by gradient boosting tree, which can reveal the complex impact of different economic indicators on land demand and enhance the prediction ability and interpretability of the model. A dynamic trend prediction model is constructed based on the Prophet model, which can take into account the continuity of time and the dynamic changes of data, thereby providing a more accurate and real-time prediction of construction land changes; the Prophet model is particularly suitable for time series data with trend, seasonal and holiday effects, and can provide interpretable prediction results. Through the prediction of dynamic stage trend data, planners can better adapt to the changes in land demand brought about by economic development and population changes, and improve the adaptability and flexibility of planning. Accurate prediction of construction land change trends helps governments and relevant departments to allocate land resources more effectively and promote sustainable economic development. Through dynamic prediction models, potential risks and uncertainties can be identified in advance, providing a more scientific basis for planning decisions and reducing risks in the planning implementation process. In summary, the effect of these steps is to provide a scientific, data-driven analysis and prediction framework that helps to achieve efficient management and reasonable planning of land resources, while providing support for sustainable urban development.

[0071] Preferably, step S2 comprises the following steps:

[0072] Step S21: setting a plurality of key influencing factors according to the dynamic stage trend data, thereby obtaining key influencing factor data, wherein the key influencing factors include economic growth rate, industrial structure adjustment and technological advancement speed;

[0073] Step S22: using the Delphi method to evaluate the possible value ranges of key influencing factor data in different scenarios, thereby obtaining factor value range data;

[0074] Step S23: constructing a simulation framework of a baseline scenario, a transformation scenario, and an innovation scenario according to the dynamic stage trend data and the factor value range data through a system dynamics model, thereby obtaining multi-scenario simulation data;

[0075] Step S24: performing Monte Carlo simulation based on the multi-scenario simulation data to generate a large number of possible development paths, thereby obtaining a scenario simulation data set;

[0076] Step S25: clustering the scenario simulation data set based on hierarchical clustering, and extracting the typical development path under each scenario, so as to obtain typical scenario path data;

[0077] Step S26: forecasting the future construction land demand based on the typical scenario path data by using a time series forecasting method, thereby obtaining multi-scenario forecasting data;

[0078] Step S27: constructing an industry-land demand sensitivity analysis model based on regional economic indicator data and historical construction land data;

[0079] Step S28: Generate possible land demand forecast ranges for different industries based on the industry-land demand sensitivity analysis model, thereby obtaining land demand elasticity forecast data.

[0080] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment of the present invention, step S2 includes the following steps:

[0081] Step S21: setting a plurality of key influencing factors according to the dynamic stage trend data, thereby obtaining key influencing factor data, wherein the key influencing factors include economic growth rate, industrial structure adjustment and technological advancement speed;

[0082] Embodiment of the present invention This is one of the core factors for predicting land demand, which directly affects the overall speed and scale of regional development. The economic growth rate can be extracted from historical data and future expectations, and its possible growth path can be set according to different scenarios; changes in industrial structure, such as the transition from manufacturing to service industry, have a significant impact on the type and quantity of land demand, and the speed and direction of industrial structure adjustment under different scenarios need to be set. Technological progress can change land use patterns, such as reducing the demand for land through more efficient land use technology, or increasing land demand through emerging industries. Therefore, this is also a key influencing factor. According to the dynamic stage trend data, set the possible development path of these key influencing factors in the future; define the value range of each factor in different scenarios (such as high growth, medium growth, low growth).

[0083] Step S22: using the Delphi method to evaluate the possible value ranges of key influencing factor data in different scenarios, thereby obtaining factor value range data;

[0084] The embodiment of the present invention selects expert group members in the field, such as economists, urban planning experts, technical experts, etc., to form a multidisciplinary expert team; through anonymous surveys, the opinions of the expert group members on the possible values ​​of each key influencing factor in different scenarios are collected. The Delphi method usually requires multiple rounds of evaluation, and each round of experts provides feedback and adjustments to the opinions of the previous round to gradually reach a consensus; finally, the opinions of the expert group are summarized to determine the value range of each key influencing factor in different scenarios, and the factor value range data is generated.

[0085] Step S23: constructing a simulation framework of a baseline scenario, a transformation scenario, and an innovation scenario according to the dynamic stage trend data and the factor value range data through a system dynamics model, thereby obtaining multi-scenario simulation data;

[0086] The embodiment of the present invention uses the System Dynamics (SD) model to establish a relationship model between land use, economic growth, industrial structure and technological progress. The SD model simulates the dynamic behavior of complex systems through feedback loops and time delays. According to the value range data of the factors, the system dynamics models under the baseline scenario (assuming that the existing trend continues), the transformation scenario (assuming that the industrial structure adjustment is accelerated) and the innovation scenario (assuming that the technological progress is significantly accelerated) are constructed respectively; in each scenario, different parameter values ​​are set, such as economic growth rate, technological progress rate, etc., to form multi-scenario simulation data.

[0087] Step S24: performing Monte Carlo simulation based on the multi-scenario simulation data to generate a large number of possible development paths, thereby obtaining a scenario simulation data set;

[0088] The Monte Carlo simulation of the embodiment of the present invention generates a large number of simulation results by randomly sampling the key parameters in the model, thereby reflecting the possible development paths and their uncertainties. Under the constructed baseline scenario, transformation scenario and innovation scenario, Monte Carlo simulation is used to generate a variety of possible development paths, which reflect the changes in the demand for construction land in the region under different assumptions. The results of all simulations are collected to form a scenario simulation data set, which contains various possible paths under each scenario and provides a basis for subsequent analysis.

[0089] Step S25: clustering the scenario simulation data set based on hierarchical clustering, and extracting the typical development path under each scenario, so as to obtain typical scenario path data;

[0090] The embodiment of the present invention uses a hierarchical clustering algorithm to group similar paths in the scenario simulation data set. The method merges similar paths layer by layer by constructing a tree-like clustering structure. The clustering results are analyzed to extract the typical development paths under each scenario. These paths represent the most likely development trends and change patterns under each scenario. The extracted typical development paths are used to form typical scenario path data for future construction land demand forecasting.

[0091] Step S26: forecasting the future construction land demand based on the typical scenario path data by using a time series forecasting method, thereby obtaining multi-scenario forecasting data;

[0092] The embodiment of the present invention selects an appropriate time series prediction model, such as a Prophet model, based on typical scenario path data to predict future changes in construction land demand. Using the selected time series model, input typical scenario path data to predict future construction land demand under different scenarios; organize the prediction results into multi-scenario prediction data to provide a reference for decision-making.

[0093] Step S27: constructing an industry-land demand sensitivity analysis model based on regional economic indicator data and historical construction land data;

[0094] The embodiment of the present invention selects a suitable sensitivity analysis method, such as regression analysis, random forest, etc., to establish a relationship model between industry and land demand; inputs regional economic indicator data, historical construction land data and multi-scenario forecast data to train and verify the model; uses the model to analyze the sensitivity of each industry to land demand, and identifies which industries' growth or adjustment has the greatest impact on land demand.

[0095] Step S28: Generate possible land demand forecast ranges for different industries based on the industry-land demand sensitivity analysis model, thereby obtaining land demand elasticity forecast data.

[0096] The embodiment of the present invention predicts the possible land demand range of different industries under different economic development scenarios based on the constructed sensitivity analysis model; based on the prediction results, generates land demand elasticity prediction data, which reflects the possible range of future land demand and its sensitivity to economic and industrial changes.

[0097] The embodiment of the present invention can more accurately simulate and predict the changing trend of construction land demand by determining key factors such as economic growth rate, industrial structure adjustment and technological progress speed, and provide a basis for policy formulation. The Delphi method is used to evaluate the possible value range of key influencing factors, which can integrate the opinions of experts in different fields and increase the reliability and authority of the prediction results. The simulation framework under different scenarios is constructed by the system dynamics model, which can evaluate the potential impact of changes in different factors on the demand for construction land and provide decision makers with a multi-angle perspective. A large number of possible development paths are generated, and by simulating uncertainty and randomness, the risks and opportunities under different development strategies can be better understood. By analyzing the scenario simulation data set through hierarchical clustering, typical development paths are extracted, which helps to identify and focus on key land use patterns and trends. The use of time series prediction methods to predict future construction land demand can provide more accurate and dynamic land demand predictions and provide a scientific basis for land supply and planning. An industry-land demand sensitivity analysis model is constructed to evaluate the impact of different industrial development on land demand and provide decision support for industrial policies and land resource allocation. Land demand elasticity prediction data can help understand the responsiveness of the land market to economic changes. In summary, the effect of these steps is to provide a comprehensive analytical and forecasting framework that not only enhances the understanding of changes in construction land demand, but also provides scientific and systematic tools and methods for urban planning, land resource management and policy making.

[0098] Preferably, step S27 includes the following steps:

[0099] Step S271: performing principal component analysis on the regional economic indicator data and extracting the main influencing factors, thereby obtaining principal component data of economic indicators;

[0100] In the embodiment of the present invention, first, the regional economic indicator data is standardized. The purpose of standardization is to eliminate the dimensional differences between different indicators so that each indicator has the same weight in the analysis; the Z standardization method with a mean of 0 and a standard deviation of 1 is usually used to convert each economic indicator into standardized data. PCA is a dimensionality reduction technique used to convert multidimensional data into a few principal components that are linearly independent and explain most of the variance of the data; the eigenvalues ​​and eigenvectors are extracted by calculating the covariance matrix of the standardized economic indicator data. The direction corresponding to the eigenvector is the direction of the principal component, and the eigenvalue represents the proportion of variance explained by the principal component; according to the proportion of variance explained, the first few principal components (for example, the first two or the first three) are selected as the principal component data of economic indicators representing the main economic influencing factors.

[0101] Step S272: classifying the land based on different industrial types according to the historical construction land data, thereby obtaining classified construction land data;

[0102] The embodiment of the present invention classifies the historical construction land data according to the types of industries in the region (such as agriculture, industry, service industry, etc.), and can be classified according to land use (such as commercial land, industrial land, agricultural land, etc.) or industrial functional zoning. The historical construction land data is aggregated by industry type to generate classified construction land data, which reflects the actual demand for land in different periods and different industries. The classified data is standardized to ensure that it can be associated with the principal component data of economic indicators for analysis.

[0103] Step S273: analyzing the correlation between the economic indicator principal component data and the classified construction land data by using the Pearson correlation coefficient, and screening the relevant indicators according to the preset correlation coefficient threshold, so as to obtain the key indicator data;

[0104] The Pearson correlation coefficient of the embodiment of the present invention is a statistical indicator that measures the degree of linear correlation between two variables, and its value range is between -1 and 1. The closer the value is to 1 or -1, the stronger the linear relationship between the two variables is; the principal component data of the economic indicators and the classified construction land data are calculated pair by pair, and the Pearson correlation coefficient between each principal component and each type of land demand is obtained. According to the preset correlation coefficient threshold (for example, 0.5 or 0.7), the principal components of the economic indicators that are highly correlated with the classified construction land data are screened out; the screened principal components are used as key indicator data, and these data are retained for subsequent model construction.

[0105] Step S274: using a multiple linear regression method with key indicator data as independent variables and classified construction land data as dependent variables, a preliminary industry-land demand relationship model is established to obtain preliminary model parameters;

[0106] The multivariate linear regression model of the embodiment of the present invention is used to study the linear relationship between the independent variable (here is the key indicator data) and the dependent variable (here is the classified construction land data); the key indicator data is used as the independent variable and the classified construction land data is used as the dependent variable to establish a regression equation. The form of the preliminary model is: Y = β0 + β1X1 + β2X2 + ... + β n X n +∈, where Y is the land demand, X i is the main component of economic indicators, β i is the regression coefficient, ∈ is the error term. The regression coefficient is estimated using the least squares method (OLS) to obtain preliminary model parameters. OLS optimizes the model by minimizing the squared error between the predicted value and the actual value.

[0107] Step S275: using a residual analysis method to identify outliers and influencing points of the preliminary model parameters, thereby obtaining model diagnostic data;

[0108] The residual of the embodiment of the present invention refers to the difference between the predicted value and the actual value of the regression model. By calculating the residual of each sample point, the fitting effect of the model can be evaluated. Using residual graphs, standardized residual graphs and other means, outliers (i.e., sample points with large residuals) and influence points (sample points with a large impact on model parameters) are identified. The identified outliers and influence points are recorded, and their impact on the model parameters is evaluated. These data will be used for further optimization and adjustment of the model.

[0109] Step S276: Adjust the preliminary model parameters according to the model diagnostic data and introduce nonlinear terms to construct an industry-land demand sensitivity analysis model.

[0110] The embodiment of the present invention decides whether to remove certain outliers or influencing points based on the results of residual analysis to improve the fitting accuracy of the model; adjusts the parameters in the preliminary model and re-estimates the regression coefficients to improve the explanatory power and predictive power of the model. In actual situations, there may be a nonlinear relationship between industrial development and land demand. Nonlinear terms (such as square terms, interaction terms, etc.) are introduced into the regression model to capture this complex relationship. Construct an industry-land demand sensitivity analysis model containing nonlinear terms. The final model may be in the form of:

[0111] The present invention performs dimensionality reduction processing on regional economic indicator data through principal component analysis, and extracts the main influencing factors reflecting economic characteristics, which helps to simplify the model and focus on key economic indicators. Based on the classified construction land data of different industrial types, the impact of various industries on land demand can be analyzed more accurately, providing segmented data support for subsequent model establishment. The Pearson correlation coefficient is used to perform correlation analysis between the principal components of economic indicators and the classified construction land data, and key indicators are screened out by preset thresholds, which helps to identify the main economic factors affecting land demand. A preliminary industry-land demand relationship model is established using the multivariate linear regression method, and model parameters are obtained, which provides a quantitative framework for understanding the relationship between industry and land demand. The model parameters are diagnosed by the residual analysis method to identify outliers and influencing points, which helps to improve the accuracy and reliability of the model. The preliminary model parameters are adjusted according to the model diagnosis data, and the industry-land demand sensitivity analysis model is constructed by introducing nonlinear terms, which enables the model to more comprehensively capture the complex relationship between industry and land demand and enhance the explanatory power and predictive ability of the model. Through these steps, a more accurate and reliable industry-land demand sensitivity analysis model can be constructed to provide scientific decision-making support for policy makers and provide a basis for the rational allocation of land resources and urban planning.

[0112] Preferably, step S28 comprises the following steps:

[0113] Step S281: extracting the industrial structure change trend under each scenario based on the multi-scenario forecast data, thereby obtaining industrial structure change data;

[0114] The embodiment of the present invention extracts the trend of industrial structure changes under each scenario based on multi-scenario forecast data, including the growth or decline of different industries under the baseline scenario, transformation scenario and innovation scenario. Use time series analysis technology to analyze key indicators such as output value or employment of different industries to identify and extract their trends over time. Use polynomial fitting or exponential smoothing methods to fit the trend of industrial structure changes, so as to obtain smoother industrial structure change data. Compare the industry change trends under each scenario to identify which industries show significant differences in different scenarios.

[0115] Step S282: setting a plurality of possible industrial development scenarios according to the industrial structure change data, thereby obtaining industrial development scenario data;

[0116] The embodiment of the present invention sets multiple possible industrial development scenarios based on the extracted industrial structure change data, each of which corresponds to different external factors such as orientation, technological progress speed, international market changes, etc. Assumptions are set for each scenario, such as different economic growth rates, industrial support, technology introduction, etc., to ensure the diversity and reality of the scenarios. Using the industrial structure change trend data, combined with the scenario assumptions, detailed industrial development scenario data are generated through simulation or expert opinions (such as the Delphi method), which include the possible development paths of each industry in different years in the future.

[0117] Step S283: inputting the industrial development scenario data into the industry-land demand sensitivity analysis model to generate preliminary land demand forecast results under each scenario, thereby obtaining preliminary demand forecast data;

[0118] The embodiment of the present invention inputs the generated industrial development scenario data into the previously constructed industry-land demand sensitivity analysis model to simulate the land demand under different scenarios. The model calculates the preliminary land demand of each industry under different scenarios based on the input scenario data. These calculation results are generated based on the previous sensitivity analysis model parameters. The preliminary land demand forecast data under each scenario are extracted. These data represent the land demand of each industry under different development scenarios, providing a basis for subsequent adjustments and uncertainty analysis.

[0119] Step S284: performing efficiency adjustment on the preliminary demand forecast data according to the basic trend data, and performing uncertainty analysis on the efficiency-adjusted demand data to perform possible demand scenarios, thereby obtaining a demand simulation data set;

[0120] The embodiment of the present invention adjusts the preliminary demand forecast data according to basic trend data (such as historical trends, technological progress, etc.), and the goal of the adjustment is to take into account the possible changes in demand brought about by efficiency improvements or changes in industrial structure. Linear regression, nonlinear regression or machine learning algorithms can be used to adjust the preliminary forecast data to make the forecast closer to the actual possible development situation. Uncertainty analysis is performed on the efficiency-adjusted data, and a large number of possible demand scenarios are generated using Monte Carlo simulation. Through repeated random sampling, a demand simulation data set is generated, covering a variety of possible future scenarios. The rationality of the demand simulation data set is verified by comparison with historical data.

[0121] Step S285: Statistically analyzing the demand simulation data set, extracting the land demand range under different confidence intervals, and thus obtaining demand interval data;

[0122] The embodiment of the present invention performs statistical analysis on the generated demand simulation data set, calculates the land demand range under different confidence intervals, for example, calculates the demand range of different confidence intervals such as 95%, 90%, and obtains the upper and lower limits of demand. These confidence interval data provide a quantitative range for the uncertainty of future land demand, thereby helping decision makers understand possible demand changes under different conditions. The results of the statistical analysis are extracted to obtain the land demand range under different confidence intervals, forming demand interval data, which provides a basis for elastic prediction.

[0123] Step S286: Construct the land demand forecast range of different industries under different development scenarios based on the industrial development scenario data and the demand range data, so as to obtain the land demand elasticity forecast data.

[0124] The embodiment of the present invention constructs a land demand elasticity prediction model based on the previously generated industrial development scenario data and demand interval data. The model can predict the range of land demand changes in various industries under different scenarios. The results of uncertainty analysis are incorporated into the model, so that it can reflect the potential fluctuation range of land demand under different scenarios and confidence intervals. Finally, the elasticity prediction data of land demand is obtained, which provides the range of land demand for different industries under various possible development scenarios. This data will be used for subsequent spatial planning and land use layout optimization.

[0125] The present invention extracts the trend of industrial structure changes under different scenarios by analyzing multi-scenario forecast data, which helps to identify the key driving factors of economic development and provide a basis for policy making. According to the industrial structure change data, multiple possible industrial development scenarios are set, which helps to evaluate the potential impact of different industrial development paths on land demand. The industrial development scenario data is input into the industry-land demand sensitivity analysis model to generate preliminary land demand forecast results under each scenario, providing a preliminary basis for the rational allocation and planning of land resources. The preliminary demand forecast data is adjusted for efficiency according to the basic trend data, and uncertainty analysis is performed, which helps to evaluate the reliability of the forecast results and provide scenario analysis for possible demand changes. The demand simulation data set is statistically analyzed to extract the land demand range under different confidence intervals, which helps to identify the uncertainty and risk of land demand and provide more comprehensive information for decision makers. According to the industrial development scenario data and the demand interval data, the land demand forecast range of different industries under different development scenarios is constructed, so as to obtain the land demand elasticity forecast data, which helps to evaluate the sensitivity of different industrial development to land demand and provide a scientific basis for the macro-control of the land market and the optimal allocation of land resources. Through these steps, a comprehensive analytical framework can be constructed, which can not only predict the changing trend of land demand, but also evaluate the potential impact under different industrial development scenarios, and provide strong decision-making support for land resource management, urban planning and policy making.

[0126] Preferably, step S3 comprises the following steps:

[0127] Step S31: using GIS technology to process and integrate the control line spatial data, thereby obtaining standardized control line layer data, wherein the data processing includes coordinate system unification and data format conversion;

[0128] The embodiment of the present invention unifies the coordinate systems of various control line data (including ecological protection red lines, permanent basic farmland lines, urban development boundaries, etc.), and usually selects a universal geographic coordinate system (such as WGS84) or a projection coordinate system (such as UTM) for unification to ensure that all data are overlaid and analyzed under the same coordinate reference. The control line data formats (such as Shapefile, GeoJSON, etc.) from different sources are uniformly converted into a consistent format for subsequent spatial analysis. The commonly used format is a vector data format (such as Shapefile) because it is more suitable for spatial analysis operations. The processed control line data is superimposed on a unified GIS platform to form multiple layers. Each layer represents a different type of control line; the layers are standardized to ensure that the boundaries, resolutions and other parameters of all control lines are consistent, thereby forming standardized control line layer data.

[0129] Step S32: performing spatial overlay analysis based on the control line layer data, and identifying the intersection area and conflict area between different control lines, thereby obtaining control line conflict data;

[0130] The embodiment of the present invention utilizes the spatial overlay analysis tool of GIS (such as the overlay analysis tool in ArcGIS) to overlay different control line layers and analyze the relationship between the control lines; the overlay analysis can identify the overlapping areas, intersection areas, and respective spatial distribution characteristics between the control lines. In the overlay analysis, special attention is paid to the conflict areas between different control lines (i.e., the intersections or overlapping areas of multiple control line layers), which may have land use conflicts and require further processing. Based on the results of the overlay analysis, the conflict data between the control lines are extracted, and the conflict data include information such as the area, location, and type of control lines involved in the overlapping area; the conflict data are classified and labeled according to the severity of the conflict (such as complete overlap, partial overlap), providing a basis for subsequent priority division and adjustment.

[0131] Step S33: performing priority division and boundary adjustment on the control line layer data based on the conflicting areas according to the control line conflict data, thereby obtaining control line adjustment data;

[0132] The embodiment of the present invention sets the priority of the control line according to the national and local land use planning. For example, the ecological protection red line usually has the highest priority, the permanent basic farmland line is second, and the urban development boundary may be at a lower priority. In the conflict area, it is determined which control line needs to be retained or adjusted in order of priority. Control lines with higher priorities usually retain their original boundaries, while those with lower priorities require boundary adjustments. Using GIS editing tools, the boundaries of low-priority control lines are adjusted to avoid conflicts with high-priority control lines. Adjustment methods may include boundary translation, reducing conflict areas, etc. The adjusted control line boundaries are saved as new layer data to form control line adjustment data. This step ensures that all control lines do not conflict with each other in space and meet planning requirements.

[0133] Step S34: spatially identifying and extracting the constructible area according to the control line adjustment data, thereby obtaining constructible area data;

[0134] The embodiment of the present invention uses control line adjustment data to perform spatial screening and identify areas that are not restricted by control lines as potential buildable areas. These areas are spaces outside all control lines and are usually used for urban expansion, industrial park construction, etc.; according to national or local planning requirements, the buildable areas are further screened. For example, avoid sensitive areas such as areas prone to geological disasters and water source protection areas. Use GIS tools to extract buildable areas and generate a buildable area layer in vector data format, which contains all regional boundary information that meets the construction conditions; integrate the extracted buildable area data into the GIS database as the basic data for subsequent land use layout optimization.

[0135] Step S35: Utilize a multi-objective optimization algorithm to optimize the land use layout according to the buildable area data and the multi-scenario prediction data, thereby obtaining optimized layout plan data.

[0136] The embodiment of the present invention sets optimization goals based on multi-scenario prediction data, such as maximizing land use efficiency, minimizing environmental impact, optimizing industrial land layout, etc.; commonly used multi-objective optimization algorithms include genetic algorithms, multi-objective particle swarm optimization algorithms (MOPSO), etc. The algorithm will search for the optimal land layout plan in the buildable area to meet multiple optimization goals. The algorithm continuously adjusts the land layout through iterative calculations and evaluates the pros and cons of different layout plans. Each iteration will generate multiple possible layout plans, which will be evaluated and screened according to the preset optimization goals. Finally, the optimal land layout plan data is extracted, which includes specific land allocation, industrial zoning, infrastructure layout and other information to meet the needs under multi-scenario prediction.

[0137] The present invention realizes the coordinate system unification and data format conversion of control line spatial data through GIS technology, ensuring the consistency and accuracy of data in subsequent analysis. This step is the basis for ensuring data quality and is crucial for subsequent spatial analysis. The spatial overlay function of GIS is used to identify the intersection and conflict areas between different control lines. This process helps to discover potential spatial planning problems and provide a basis for subsequent planning adjustments. According to the conflict identification results, the control line layer is prioritized and the boundary is adjusted. This helps to solve spatial conflict problems, optimize the control line layout, and ensure the rationality of land use planning. Based on the adjusted control line layer, the buildable area is identified and extracted. This step provides accurate spatial information for land use planning and construction activities, ensuring the rational allocation and utilization of land resources. Combined with multi-scenario prediction data and buildable area data, a multi-objective optimization algorithm is used to optimize the land use layout. This optimization method can balance different land use needs, achieve efficient use of land resources, and support the formation of a scientific and reasonable land use planning scheme. Through these steps, the control line spatial data can be effectively processed and analyzed, spatial conflict problems can be solved, and scientific and reasonable decision support can be provided for land use planning and construction activities. The application of GIS technology plays a core role in this process. It not only improves the efficiency of data processing, but also enhances the scientificity and accuracy of planning decisions.

[0138] Preferably, step S35 includes the following steps:

[0139] Step S351: Conducting a suitability assessment on the constructible area data based on natural conditions, and screening out areas suitable for construction, thereby obtaining construction suitability data;

[0140] The embodiment of the present invention collects natural condition data that affect construction suitability, including terrain slope, soil type, hydrological conditions, geological disaster risks, climate conditions, etc. A construction suitability assessment model is constructed using a weighted comprehensive scoring method or a hierarchical analysis method (AHP), a weight is assigned to each natural condition, and a suitability score is calculated based on the natural conditions of a specific area; the calculated suitability score is classified, usually into highly suitable, moderately suitable, and unsuitable categories; based on the suitability classification, areas suitable for construction are screened out to generate construction suitability data. These data include the suitability level of each area and its spatial distribution information.

[0141] Step S352: obtaining existing construction land distribution data; performing spatial continuity and concentration analysis based on the construction suitability data and the existing construction land distribution data, thereby obtaining spatial layout optimization data;

[0142] The embodiment of the present invention obtains the construction land distribution data of the currently developed area, including the specific location and area of ​​residential areas, industrial areas, commercial areas, etc.; uses GIS technology to perform spatial continuity analysis to evaluate whether the distribution of existing construction land forms a continuous development or is scattered. Calculate land concentration indicators, such as the distribution balance and agglomeration of land use, and use methods such as Moran's I index or Gini coefficient to measure whether the spatial distribution of existing construction land is concentrated; identify areas that need to be enhanced or adjusted based on the results of spatial continuity and concentration analysis, and generate spatial layout optimization data. These data are used to guide subsequent land layout optimization.

[0143] Step S353: spatially allocating land use requirements under different scenarios according to the multi-scenario prediction data, generating a preliminary land use layout plan, and thus obtaining preliminary layout plan data;

[0144] The embodiment of the present invention uses the multi-scenario prediction data in step S26 to input the land demand under different scenarios in the future; selects a suitable spatial allocation method, such as a gravity model or a rule-based allocation model, and allocates land according to the prediction results of different scenarios. The land demand data of each scenario is spatially allocated according to different types of construction land (such as residential, industrial, commercial, etc.), taking into account the demand and spatial suitability of each type of land; generates a preliminary land layout plan, including the distribution, area and location of various types of construction land, and forms preliminary layout plan data.

[0145] Step S354: Utilize a multi-objective optimization algorithm to optimize the preliminary layout plan data based on land use efficiency, ecological environment protection, and traffic accessibility goals with the spatial layout optimization data as constraints, and perform spatial conflict detection and local adjustments based on land use conflicts to obtain optimized layout plan data.

[0146] The embodiment of the present invention determines the objectives of multi-objective optimization, including maximizing land use efficiency, optimizing ecological environment protection, optimizing traffic accessibility, etc. Each objective will have a corresponding weight; select a suitable multi-objective optimization algorithm, such as genetic algorithm, multi-objective particle swarm optimization (MOPSO), etc., to perform optimization calculations. The preliminary layout plan data and the spatial layout optimization data are used as inputs, and iterative optimization is performed through the multi-objective optimization algorithm to adjust the layout of various types of land to meet the set multi-objective requirements; during the optimization process, GIS tools are used to detect spatial conflicts and identify conflicting areas between the layouts of various types of land, such as spatial overlap or unreasonable distance arrangement between residential areas and industrial areas. According to the detected spatial conflicts, local adjustments are made to ensure that the layouts of various types of land are reasonably distributed and the conflicting areas are reduced or eliminated; finally, the optimized layout plan data is generated, covering the reasonable layout of various types of land, and maximizing the goals of land use efficiency, ecological environment protection and traffic accessibility. These data will serve as the final basis for construction planning and land use decisions.

[0147] The present invention can screen out suitable areas for construction through suitability assessment based on natural conditions, ensure that construction projects can be carried out in areas with better natural conditions, thereby improving the feasibility and success rate of construction and reducing the negative impact on the environment. Obtaining existing construction land distribution data, combining construction suitability data to perform spatial continuity and concentration analysis, helps to identify the spatial distribution characteristics and potential expansion directions of construction land, and provides a scientific basis for the rational allocation and planning of land resources. According to multi-scenario prediction data, the land demand under different scenarios is spatially allocated to generate a preliminary land layout plan, which helps decision makers evaluate the impact of different development scenarios on land demand, so as to make more reasonable planning decisions. Using a multi-objective optimization algorithm, combined with goals such as land use efficiency, ecological environment protection, and traffic accessibility, the preliminary layout plan is optimized, and through spatial conflict detection and local adjustment of land use conflicts, the optimized layout plan data is obtained. This method can balance the relationship between different goals, achieve the optimal configuration of land use, and ensure the sustainable development of the ecological environment. Through these steps, land use planning can be effectively carried out to ensure that land resources are reasonably used, while taking into account factors such as economic development, environmental protection, and social needs, to achieve the scientificity and sustainability of land use planning.

[0148] Preferably, step S4 comprises the following steps:

[0149] Step S41: unifying the data format and scale of the multi-scenario prediction data, the land demand elasticity prediction data, and the optimized layout scheme data, thereby obtaining a standardized prediction data set;

[0150] The embodiment of the present invention converts the format of multi-scenario prediction data, land demand elasticity prediction data, and optimized layout scheme data to conform to a unified standard data format (such as CSV, GeoJSON, etc.). The spatial and temporal scales of the data are uniformly adjusted. For example, if some data are in annual units and other data are quarterly, they need to be unified into the same time scale; at the same time, the spatial scale needs to ensure that all data are expressed at the same geographic resolution. Data from different sources are numerically standardized to eliminate dimensional differences; for example, the data can be converted to a standard distribution between 0 and 1 or the Z-score normalization method can be used. The data with unified format and scale are integrated to generate a standardized prediction data set; this data set will be used for the subsequent construction and training of the Bayesian network model.

[0151] Step S42: constructing a Bayesian network structure according to the standardized prediction data set, defining the causal relationship between nodes and the conditional probability distribution, thereby obtaining an initial Bayesian network model; using historical construction land data to perform parameter learning on the initial Bayesian network model, optimizing the conditional probability table, thereby obtaining a trained Bayesian network model;

[0152] The embodiment of the present invention defines nodes in the Bayesian network based on key variables in the standardized prediction data set; for example, the nodes may include construction land demand, economic growth, industrial structure, etc. Expert knowledge or structural learning algorithms (such as PC algorithm or MMHC algorithm) are used to define the causal relationship between nodes and determine the network structure. The conditional probability distribution (CPT) of each node is preliminarily set, and the initial conditional probability table is defined according to the relationship between the variables. The initial Bayesian network model is parameter learned using historical construction land data to optimize the conditional probability table; learning methods such as the expectation maximization (EM) algorithm are used to adjust the conditional probability in the network to best fit the historical data; the accuracy and reliability of the trained Bayesian network model are verified by methods such as cross-validation or posterior prediction checks.

[0153] Step S43: input the standardized prediction data set into the trained Bayesian network model, perform inference calculation, obtain the posterior probability distribution of each node, and thus obtain comprehensive prediction result data;

[0154] The embodiment of the present invention inputs the standardized prediction data set into the trained Bayesian network model to perform inference calculation; uses Bayesian reasoning to update the posterior probability distribution of each node in the network according to the input data to generate comprehensive prediction result data; the posterior probability reflects the possibility of each variable under given input conditions.

[0155] Step S44: performing multiple random sampling and simulations on the comprehensive prediction result data by using the Monte Carlo simulation method to generate possible prediction results, thereby obtaining a simulation result data set;

[0156] The embodiment of the present invention uses the Monte Carlo simulation method to perform a large amount of random sampling on the comprehensive prediction result data to simulate possible results under different situations; through multiple simulations, a series of possible prediction results are generated, and these results constitute a simulation result data set. The more simulation times, the more statistically significant the results are.

[0157] Step S45: Statistically analyzing the simulation result data set, calculating the confidence interval and uncertainty range of the prediction result, and thus obtaining uncertainty assessment data;

[0158] The embodiment of the present invention performs statistical analysis on the simulation result data set and calculates the confidence interval of the prediction result; the confidence interval provides the possible range of the prediction result under a certain confidence level; the uncertainty of the prediction result is evaluated, including deterministic and variability analysis, to determine the fluctuation range of the prediction result under different scenarios.

[0159] Step S46: construct the final elastic prediction result data based on the comprehensive prediction result data and the uncertainty assessment data, wherein the elastic prediction result data includes the predicted median, confidence interval and possible fluctuation range.

[0160] The embodiment of the present invention integrates the comprehensive prediction result data with the uncertainty assessment data to generate the final elastic prediction result data. The elastic prediction result data includes the median prediction value, the confidence interval, and the possible fluctuation range; these data provide a scientific basis for the planning and management of construction land demand and reflect the impact of future uncertainty.

[0161] The present invention ensures that data from different sources and types can be compared and analyzed under the same framework by unifying the data format and scale, and provides high-quality input data sets for subsequent model construction and prediction. The causal relationship and conditional probability distribution between nodes are defined to construct an initial Bayesian network model. The model is parameterized using historical construction land data, the conditional probability table is optimized, and the accuracy and prediction ability of the model are improved. The standardized prediction data set is input into the trained Bayesian network model, and the reasoning calculation is performed to obtain the posterior probability distribution of each node, which helps to understand the dependency between different variables and provide probabilistic support for decision-making. The comprehensive prediction results are randomly sampled and simulated multiple times by the Monte Carlo simulation method to generate possible prediction results, which helps to evaluate the distribution characteristics and potential risks of the prediction results. The simulation result data set is statistically analyzed to calculate the confidence interval and uncertainty range of the prediction results, providing decision makers with important information on the stability and reliability of the prediction results. Combined with the comprehensive prediction results and uncertainty assessment data, the final elastic prediction result data is constructed, including the predicted median, confidence interval and possible fluctuation range, providing solid data support for long-term planning and strategy formulation. Through these steps, a prediction model that comprehensively considers multiple factors and uncertainties can be constructed, providing a scientific, flexible and resilient decision-making basis for land use planning and the formulation of related policies.

[0162] Preferably, step S5 comprises the following steps:

[0163] Step S51: Acquire planning indicator data of different planning scales;

[0164] The embodiment of the present invention determines the planning indicators required at different planning scales (such as regional level, city level, community level), including land use type distribution, population density, transportation network density, environmental quality, etc.; collects relevant planning indicator data from government planning departments, statistical yearbooks, field surveys and other channels. The collected planning indicator data are sorted, including unifying the data format, filling missing values ​​and processing abnormal values; the indicator data at different scales are standardized to ensure that the data is compared and analyzed under the same dimension.

[0165] Step S52: extracting the land demand forecast results under different scenarios according to the elastic forecast result data, thereby obtaining the scenario-based forecast data;

[0166] The embodiment of the present invention extracts land demand forecast results under different scenarios (such as baseline scenario, transformation scenario, and innovation scenario) from the elastic forecast result data; extracts information such as land demand, land type distribution, and spatial layout under each scenario to form scenario-specific forecast data.

[0167] Step S53: establishing a hierarchical relationship between indicators according to the planning indicator data, thereby obtaining indicator hierarchical structure data; using a fuzzy comprehensive evaluation method to evaluate the importance of each indicator in the indicator hierarchical structure data, thereby obtaining indicator weight data;

[0168] The embodiment of the present invention establishes a hierarchical relationship between indicators according to different planning scales and planning goals, for example, a hierarchical structure from overall goals to sub-goals and then to specific indicators is established; indicator hierarchical structure data is constructed, including the position of each indicator in the hierarchy and their mutual relationship. An appropriate fuzzy comprehensive evaluation method (such as fuzzy AHP, fuzzy hierarchical analysis method, etc.) is selected to evaluate the relative importance of the indicators; the relative importance scores of the indicators are obtained through expert scoring or questionnaire surveys, and converted into weight data.

[0169] Step S54: construct a multi-objective planning model based on the indicator weight data and the scenario-based prediction data, with the goal of minimizing the deviation between the planning indicators of different scales and the prediction results, thereby obtaining a preliminary coordination model;

[0170] The embodiment of the present invention constructs a multi-objective planning model based on the indicator weight data and the scenario-based prediction data, wherein the objectives include minimizing the deviation between the planning indicators and the prediction results at different scales; setting an optimization objective function, for example, minimizing the prediction deviation of each planning scale by weighted square error. Using linear programming or nonlinear programming technology, a preliminary coordination model is constructed to reflect the relationship between land use demand and planning indicators at different scales.

[0171] Step S55: solving the preliminary coordination model to generate preliminary coordination results at each planning scale, thereby obtaining preliminary coordination data;

[0172] The embodiment of the present invention selects a suitable optimization algorithm (such as interior point method, genetic algorithm, simulated annealing, etc.) to solve the preliminary coordination model; solves the model to generate preliminary coordination results at each planning scale, and obtains preliminary coordination data, including the matching status of land allocation scheme and planning indicators.

[0173] Step S56: Perform consistency check on the preliminary coordination data to identify possible contradictions and conflicts between different planning scales, thereby obtaining coordination conflict data;

[0174] The embodiment of the present invention performs consistency check on the preliminary coordination data and planning indicators to identify possible contradictions and conflicts between different planning scales; analyzes the deviation between the land use demand and planning indicators at each scale to identify potential planning conflict areas or inconsistencies. The conflicting data found in the consistency check is recorded, including specific indicator conflict areas and data contradictions.

[0175] Step S57: using a heuristic algorithm to locally adjust and optimize the preliminary coordination data based on the coordination conflict data, and balancing the requirements of different planning scales, thereby obtaining optimized coordination data;

[0176] The embodiment of the present invention selects a heuristic algorithm (such as a genetic algorithm, a simulated annealing algorithm, etc.) for local adjustment and optimization; according to the coordination conflict data, the preliminary coordination data is locally adjusted to balance the needs of different planning scales and ensure the coordination and consistency of each scale; multiple iterative optimizations are performed to adjust the land allocation and planning indicators at each scale to eliminate conflicts and achieve a balance among various goals; and optimized coordination data is generated to ensure the rationality and consistency of the planning scheme.

[0177] Step S58: construct a final multi-scale land demand coordination model based on the optimized coordination data and the elastic prediction result data, and output the coordinated land scale prediction data.

[0178] The embodiment of the present invention combines the optimized coordination data with the elastic prediction result data to construct the final multi-scale land demand coordination model; output the final coordinated land scale prediction data, including the land allocation scheme and planning indicator coordination results of each planning scale. The final prediction data is verified to ensure its accuracy and reliability; the final results are used for actual national space planning and land use decision-making to guide future land planning and management work.

[0179] The present invention ensures the compatibility of data collected from different sources and scales, eliminates data inconsistencies, and lays a solid foundation for subsequent analysis. The standardization process improves data quality and reduces analysis errors caused by data format problems. It provides an in-depth understanding of land demand under different development scenarios, allowing decision makers to evaluate the potential impact of different scenarios on land use, enhancing the adaptability and flexibility of planning, and enabling planning to respond to different economic development paths. An indicator system that comprehensively considers the interests and goals of multiple parties has been established to ensure the multi-dimensionality and balance of planning. The weight distribution reflects the relative importance of different indicators in planning decisions, enhancing the pertinence and effectiveness of planning. It allows multiple planning goals, such as economic development, environmental protection and social needs, to be considered simultaneously, to achieve multi-objective coordination and optimization, and to ensure the optimality of planning schemes under multiple constraints through mathematical models. The solution process provides a quantitative framework for evaluating and comparing the efficiency and effectiveness of different planning schemes. The preliminary results provide a benchmark for further analysis and adjustment, and help identify potential problems in planning. Ensure the consistency of planning schemes at different scales and levels, reduce conflicts and contradictions in the process of planning implementation, and through conflict identification, potential problems in planning can be discovered and solved in advance, thereby improving the stability and reliability of planning. Applying heuristic algorithms to make local adjustments to planning schemes improves the adaptability and response speed of planning. The optimization process takes into account the constraints and limitations of actual operations, enhancing the feasibility of planning schemes. It integrates multi-scale, multi-scenario and multi-objective land use needs, provides a comprehensive planning scheme, and the output coordinated land scale forecast data provides a scientific basis for the rational allocation of land resources and planning implementation, which helps to achieve the sustainability and efficiency of land use. These steps together constitute a systematic and scientific planning process, which not only improves the quality and efficiency of planning, but also provides flexible strategies and tools for coping with complex and changing planning environments. Through this process, decision makers can better understand how various factors affect land use and formulate more reasonable, effective and sustainable planning schemes.

[0180] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0181] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for predicting the scale of construction land based on national land space planning, characterized in that: The following steps are involved: Step S1: Obtain historical construction land data, regional economic indicator data, and control line spatial data of different regions; identify potential development turning points based on historical construction land data and regional economic indicator data, and perform stage trend analysis to obtain dynamic stage trend data; Step S2: Perform multi-scenario forecasting based on the dynamic stage trend data, thereby obtaining multi-scenario forecasting data, wherein the multi-scenario includes a baseline scenario, a transformation scenario, and an innovation scenario; Construct an industry-land demand sensitivity analysis model based on regional economic indicator data and historical construction land data; generate possible land demand forecast ranges for different industries based on the industry-land demand sensitivity analysis model, thereby obtaining land demand elasticity forecast data; Step S3: Use GIS technology to perform spatial analysis on the available land under the constraints of the control line spatial data, so as to obtain the data of the buildable area; use a multi-objective optimization algorithm to optimize the land use layout according to the buildable area data and multi-scenario prediction data, so as to obtain the optimized layout plan data; Step S4: integrating the multi-scenario prediction data, the land demand elasticity prediction data and the optimized layout scheme data based on the Bayesian network model, and performing uncertainty assessment to obtain elasticity prediction result data; Step S5: Acquire planning indicator data of different planning scales; construct a multi-scale land demand coordination model based on the elastic prediction result data and the planning indicator data, so as to obtain coordinated land scale prediction data.

2. The method for predicting the scale of construction land based on national land space planning according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Obtain historical construction land data, including the area, distribution and changes of different types of construction land; Step S12: Collecting regional GDP, fixed asset investment, industrial structure ratio, population size and composition, and aligning them with historical construction land data to obtain regional economic indicator data; Step S13: Acquire spatial data of control lines in different regions, including GIS layer data of ecological protection red lines, permanent basic farmland lines, and urban development boundaries; Step S14: cleaning and standardizing the historical construction land data and regional economic indicator data to remove abnormal values ​​and missing values, thereby obtaining standard construction land data; Step S15: using time series decomposition technology to separate the trend, seasonality and random components of the construction land standard data, so as to obtain the basic trend data of the construction land change; Step S16: Identify potential development turning points of the basic trend data and perform stage trend analysis to obtain dynamic stage trend data.

3. The method for predicting the scale of construction land based on national land space planning according to claim 2 is characterized in that: Step S16 includes the following steps: Step S161: using the Bayesian change point detection method to identify potential development turning points of basic trend data, thereby obtaining key turning point data; Step S162: dividing the basic trend data into different development stages according to the key turning point data, and extracting features for each stage, thereby obtaining stage feature data; Step S163: using a gradient boosting tree to perform a nonlinear relationship analysis between regional economic indicators and changes in construction land according to stage characteristic data, thereby obtaining indicator impact weight data; Step S164: constructing a dynamic trend prediction model based on the Prophet model according to the stage characteristic data and the indicator influence weight data; Step S165: Use the dynamic trend prediction model to make a preliminary prediction of the trend of construction land changes in a certain period in the future, so as to obtain dynamic stage trend data.

4. The method for predicting the scale of construction land based on national land space planning according to claim 3 is characterized in that: Step S2 includes the following steps: Step S21: setting a plurality of key influencing factors according to the dynamic stage trend data, thereby obtaining key influencing factor data, wherein the key influencing factors include economic growth rate, industrial structure adjustment and technological advancement speed; Step S22: using the Delphi method to evaluate the possible value ranges of key influencing factor data in different scenarios, thereby obtaining factor value range data; Step S23: constructing a simulation framework of a baseline scenario, a transformation scenario, and an innovation scenario according to the dynamic stage trend data and the factor value range data through a system dynamics model, thereby obtaining multi-scenario simulation data; Step S24: performing Monte Carlo simulation based on the multi-scenario simulation data to generate a large number of possible development paths, thereby obtaining a scenario simulation data set; Step S25: clustering the scenario simulation data set based on hierarchical clustering, and extracting the typical development path under each scenario, so as to obtain typical scenario path data; Step S26: forecasting the future construction land demand based on the typical scenario path data by using a time series forecasting method, thereby obtaining multi-scenario forecasting data; Step S27: constructing an industry-land demand sensitivity analysis model based on regional economic indicator data and historical construction land data; Step S28: Generate possible land demand forecast ranges for different industries based on the industry-land demand sensitivity analysis model, thereby obtaining land demand elasticity forecast data.

5. The method for predicting the scale of construction land based on national land space planning according to claim 4 is characterized in that: Step S27 includes the following steps: Step S271: performing principal component analysis on the regional economic indicator data and extracting the main influencing factors, thereby obtaining principal component data of economic indicators; Step S272: classifying the land based on different industrial types according to the historical construction land data, thereby obtaining classified construction land data; Step S273: analyzing the correlation between the economic indicator principal component data and the classified construction land data by using the Pearson correlation coefficient, and screening the relevant indicators according to the preset correlation coefficient threshold, so as to obtain the key indicator data; Step S274: using a multiple linear regression method with key indicator data as independent variables and classified construction land data as dependent variables, a preliminary industry-land demand relationship model is established to obtain preliminary model parameters; Step S275: using a residual analysis method to identify outliers and influence points of the preliminary model parameters, thereby obtaining model diagnostic data; Step S276: Adjust the preliminary model parameters according to the model diagnostic data and introduce nonlinear terms to construct an industry-land demand sensitivity analysis model.

6. The method for predicting the scale of construction land based on national land space planning according to claim 5 is characterized in that: Step S28 includes the following steps: Step S281: extracting the industrial structure change trend under each scenario based on the multi-scenario forecast data, thereby obtaining industrial structure change data; Step S282: setting a plurality of possible industrial development scenarios according to the industrial structure change data, thereby obtaining industrial development scenario data; Step S283: inputting the industrial development scenario data into the industry-land demand sensitivity analysis model to generate preliminary land demand forecast results under each scenario, thereby obtaining preliminary demand forecast data; Step S284: performing efficiency adjustment on the preliminary demand forecast data according to the basic trend data, and performing uncertainty analysis on the efficiency-adjusted demand data to perform possible demand scenarios, thereby obtaining a demand simulation data set; Step S285: Statistically analyzing the demand simulation data set, extracting the land demand range under different confidence intervals, and thus obtaining demand interval data; Step S286: Construct the land demand forecast range of different industries under different development scenarios based on the industrial development scenario data and the demand range data, so as to obtain the land demand elasticity forecast data.

7. The method for predicting the scale of construction land based on national land space planning according to claim 6 is characterized in that: Step S3 includes the following steps: Step S31: using GIS technology to process and integrate the control line spatial data, thereby obtaining standardized control line layer data, wherein the data processing includes coordinate system unification and data format conversion; Step S32: performing spatial overlay analysis based on the control line layer data, and identifying the intersection area and conflict area between different control lines, thereby obtaining control line conflict data; Step S33: performing priority division and boundary adjustment on the control line layer data based on the conflicting areas according to the control line conflict data, thereby obtaining control line adjustment data; Step S34: spatially identifying and extracting the constructible area according to the control line adjustment data, thereby obtaining constructible area data; Step S35: Utilize a multi-objective optimization algorithm to optimize the land use layout according to the buildable area data and the multi-scenario prediction data, thereby obtaining optimized layout plan data.

8. The method for predicting the scale of construction land based on national land space planning according to claim 7 is characterized in that: Step S35 includes the following steps: Step S351: Conduct a suitability assessment on the constructible area data based on natural conditions, and select areas suitable for construction, thereby obtaining construction suitability data; Step S352: obtaining existing construction land distribution data; performing spatial continuity and concentration analysis based on the construction suitability data and the existing construction land distribution data, thereby obtaining spatial layout optimization data; Step S353: spatially allocate the land use requirements under different scenarios according to the multi-scenario prediction data, generate a preliminary land use layout plan, and thus obtain preliminary layout plan data; Step S354: Utilize a multi-objective optimization algorithm to optimize the preliminary layout plan data based on land use efficiency, ecological environment protection, and traffic accessibility goals with the spatial layout optimization data as constraints, and perform spatial conflict detection and local adjustments based on land use conflicts to obtain optimized layout plan data.

9. The method for predicting the scale of construction land based on national land space planning according to claim 8, characterized in that: Step S4 includes the following steps: Step S41: unifying the data format and scale of the multi-scenario prediction data, the land demand elasticity prediction data, and the optimized layout scheme data, thereby obtaining a standardized prediction data set; Step S42: constructing a Bayesian network structure according to the standardized prediction data set, defining the causal relationship between nodes and the conditional probability distribution, thereby obtaining an initial Bayesian network model; using historical construction land data to perform parameter learning on the initial Bayesian network model, optimizing the conditional probability table, thereby obtaining a trained Bayesian network model; Step S43: input the standardized prediction data set into the trained Bayesian network model, perform inference calculation, obtain the posterior probability distribution of each node, and thus obtain comprehensive prediction result data; Step S44: performing multiple random sampling and simulations on the comprehensive prediction result data by using the Monte Carlo simulation method to generate possible prediction results, thereby obtaining a simulation result data set; Step S45: Statistically analyzing the simulation result data set, calculating the confidence interval and uncertainty range of the prediction result, and thus obtaining uncertainty assessment data; Step S46: construct the final elastic prediction result data based on the comprehensive prediction result data and the uncertainty assessment data, wherein the elastic prediction result data includes the predicted median, confidence interval and possible fluctuation range.

10. The method for predicting the scale of construction land based on national land space planning according to claim 9, characterized in that: Step S5 includes the following steps: Step S51: Acquire planning indicator data of different planning scales; Step S52: extracting the land demand forecast results under different scenarios according to the elastic forecast result data, thereby obtaining the scenario-based forecast data; Step S53: establishing a hierarchical relationship between indicators according to the planning indicator data, thereby obtaining indicator hierarchical structure data; using a fuzzy comprehensive evaluation method to evaluate the importance of each indicator in the indicator hierarchical structure data, thereby obtaining indicator weight data; Step S54: construct a multi-objective planning model based on the indicator weight data and the scenario-based prediction data, with the goal of minimizing the deviation between the planning indicators of different scales and the prediction results, thereby obtaining a preliminary coordination model; Step S55: solving the preliminary coordination model to generate preliminary coordination results at each planning scale, thereby obtaining preliminary coordination data; Step S56: Perform consistency check on the preliminary coordination data to identify possible contradictions and conflicts between different planning scales, thereby obtaining coordination conflict data; Step S57: using a heuristic algorithm to locally adjust and optimize the preliminary coordination data based on the coordination conflict data, and balancing the requirements of different planning scales, thereby obtaining optimized coordination data; Step S58: construct a final multi-scale land demand coordination model based on the optimized coordination data and the elastic prediction result data, and output the coordinated land scale prediction data.

Images