National land space planning method based on deep learning

Abstract

The present invention relates to the technical field of land space planning, which discloses a land space planning method based on deep learning, including the steps of target division, data collection, feature analysis, comprehensive analysis, and planning judgment; by dividing the target area into a grid, and collecting and analyzing different feature data, a refined evaluation of the land space can be achieved; by calculating the average value, standard deviation, etc., the degree of discreteness of the feature data in the target area can be effectively evaluated. This helps to discover the abnormal values ​​and distribution patterns of the data, and provides a more reliable basis for subsequent planning judgments; the comprehensive impact of multiple feature data is considered, and a comprehensive evaluation index is calculated. This makes the evaluation of the target area more comprehensive and integrated, avoiding the limitations of single feature data evaluation.

Description

Technical Field

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

[0002] With the rapid development of social economy and the continuous advancement of urbanization, reasonable land space planning has become crucial. Traditional land space planning methods often rely on limited data and subjective experience judgments, lack of comprehensive, in-depth and accurate analysis of regional characteristics, resulting in problems such as insufficient scientificity of planning schemes, unreasonable resource allocation, and ineffective ecological and environmental protection.

[0003] In today's digital age, large amounts of data and advanced technologies have provided new opportunities and challenges for national space planning. On the one hand, the rich geographical, ecological, economic and other data have made the understanding of national space more comprehensive and detailed; on the other hand, how to effectively integrate and analyze these multi-source heterogeneous data and extract valuable information to support scientific and reasonable planning decisions has become a key issue that needs to be solved urgently.

[0004] Especially when faced with complex land use needs and ecological environmental protection requirements, traditional planning methods find it difficult to accurately assess the impact of various characteristics such as land fertility, topography, and hydrology on spatial planning areas, and it is difficult to achieve optimal allocation and sustainable use of land resources.

[0005] Therefore, the present invention aims to propose a national land space planning method based on deep learning, which can realize the refined and scientific planning of national land space through in-depth mining and precise analysis of multi-source data to meet the dual needs of social and economic development and ecological environmental protection. Summary of the invention

[0006] The purpose of the present invention is to provide a national land space planning method based on deep learning, which solves the technical problems raised in the background technology.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] The national land space planning method based on deep learning includes the following steps:

[0009] Step 1: Target Division

[0010] Divide the target area into a grid and obtain several unit areas and corresponding grid nodes;

[0011] Step 2: Data Collection

[0012] At the grid nodes corresponding to the target area, the characteristic data of the corresponding locations are collected. The characteristic data include soil characteristics, terrain characteristics, and hydrological characteristics:

[0013] Step 3: Feature Analysis

[0014] Obtain the quantized values ​​of the feature data corresponding to all grid nodes, and analyze the feature index of the target area based on it;

[0015] Step 4: Comprehensive Analysis

[0016] Calculate the comprehensive evaluation index of the target area according to the characteristic indexes corresponding to the multiple characteristic data;

[0017] Step 5: Planning and determination

[0018] According to the characteristic index of the target area, it is compared with the predetermined index thresholds corresponding to several pre-classified spatial planning areas, and based on the comparison and judgment results, a display form containing planning recommendations and planning avoidance corresponding spatial planning areas is obtained:

[0019] Among them, spatial planning areas refer to areas with different pre-planned functions, including agricultural development areas, ecological functional areas, urban development areas, mineral resource development areas, and water resource protection areas. Planning recommendation refers to the suggestion to set the target area as a certain spatial planning area, and planning avoidance refers to the suggestion not to set the target area as a certain spatial planning area.

[0020] As a further solution of the present invention: in step 3, the analysis method of the characteristic index corresponding to each characteristic data is the same, and the soil characteristics are selected, and the analysis of the characteristic index is as follows:

[0021] Step D1, obtain the soil fertility index corresponding to each grid node, and use it as the quantitative value of the characteristic data, that is, the quantitative value of the soil characteristic, and then mark it as Tj, j = 1, 2, ... m, m represents the number of grid nodes;

[0022] Step D2, calculate the average value of Tj of all grid nodes in the target area and record it as T0p;

[0023] Step D3, calculate the standard deviation of Tj of all grid nodes in the target area through the standard deviation formula and mark it as Tb;

[0024] Among them, the standard deviation Tb represents the degree of dispersion of the quantized value of the characteristic data in the target area;

[0025] Step D4, compare the standard deviation Tb with the preset discrete threshold Ty:

[0026] If Tb>Ty, it means that the discrete degree of the quantized value of the characteristic data in the target area is large;

[0027] Then, the corresponding Tj values ​​are deleted in descending order according to |Tj-T0p| and the standard deviation Tb of the remaining feature data quantization values ​​is calculated accordingly until Tb≤Ty;

[0028] Then, when Tb≤Ty, the number u of deleted Tj is obtained, and the average value of undeleted Tj is calculated and marked as Tpp;

[0029] Then, the corresponding quantity proportion B is calculated through B= (mu) / m;

[0030] Among them, when comparing: if Tb≤Ty, it means that the discrete degree of the quantized value of the characteristic data in the target area is moderate, then the B value is 1, and Tpp=T0p;

[0031] Step D5, through the formula: , calculate the characteristic index TZ of the corresponding characteristic data of the target area;

[0032] Among them, α1, α2, and α3 are all preset weight coefficients, μ is the correction factor, and the value of μ is 0.65;

[0033] As a further solution of the present invention: the calculation formula of the comprehensive evaluation index is:

[0034] ;

[0035] Where Z0 represents the comprehensive evaluation index, TZk represents the characteristic index corresponding to each characteristic data, k=1, 2, ... t, t represents the number of characteristic data, and TZp is the average value of multiple characteristic data TZk;

[0036] As a further solution of the present invention: the comparison and determination method is as follows:

[0037] StepK1, obtain the index threshold Z0y corresponding to the comprehensive evaluation index in each spatial planning area;

[0038] Step K2, then compare the comprehensive evaluation index Z0 of the target area with each index threshold Z0y:

[0039] If Z0≥Z0y, the spatial planning area is marked as reserved, and the corresponding comprehensive index difference Z0C of the corresponding spatial planning area is calculated by Z0C=Z0-Z0y;

[0040] If Z0<Z0y, the spatial planning area will not be marked as reserved;

[0041] Step K3, obtain all spatial planning areas with reservation marks, and recommend the corresponding spatial planning areas as the planning of the target area, and all spatial planning areas without reservation marks as planning avoidance of the target area;

[0042] Step K4: According to the index difference from large to small, the spatial planning areas recommended for planning and avoided for planning are classified and sorted, and a display form is generated.

[0043] As a further solution of the present invention: the soil characteristic is a quantitative value of soil fertility, which is divided into several fertility levels according to the measurement results of soil fertility, and a predetermined score is assigned to each fertility level;

[0044] When the soil fertility index at the corresponding location is at a certain fertility level, the score assigned to the fertility level is used as the quantitative value of the soil characteristics.

[0045] As a further solution of the present invention: wherein, soil fertility is calculated based on the quantified values ​​of multiple indicator data and the preset weight coefficients corresponding to the indicator data:

[0046] The indicator data include organic matter, pH, soil moisture content, invertase, and respiration intensity;

[0047] The formula is: L=Fi×Wi, i=1, 2, 3,...n, where n represents the number of corresponding indicator data when calculating the parameter L value, L represents the soil fertility index, Fi is the quantitative value of the i-th indicator data, and Wi is the weight coefficient of the i-th indicator data.

[0048] As a further solution of the present invention: the terrain feature is quantified by altitude, which is divided into several specific altitude intervals, and different altitude intervals are assigned predetermined scores;

[0049] When the altitude of the corresponding location is within a certain altitude range, the score assigned to the altitude range is used as the quantitative value of the terrain feature.

[0050] As a further solution of the present invention: the hydrological characteristic uses the groundwater level as a quantitative value, and sets a number of difference intervals according to a predetermined reasonable range of the groundwater level and the difference corresponding to the groundwater level exceeding the reasonable range, and assigns a predetermined score to each difference interval, wherein the reasonable range is also assigned a corresponding predetermined score;

[0051] When the groundwater level at the corresponding location is within a reasonable range, the score assigned to the reasonable range is used as the quantitative value of the hydrological characteristic.

[0052] As a further solution of the present invention: when the groundwater level at the corresponding position is not within a reasonable range, the groundwater level is first compared with the reasonable range:

[0053] If the groundwater level is lower than the reasonable range, the absolute value of the difference between the groundwater level and the minimum value in the reasonable range is calculated, and then the score assigned to the difference interval corresponding to the absolute value of the difference is used as the quantitative value of the hydrological characteristic;

[0054] If the groundwater level is higher than a reasonable range, the absolute value of the difference between the groundwater level and the maximum value in the reasonable range is calculated, and then the score assigned to the difference interval corresponding to the absolute value of the difference is used as the quantitative value of the hydrological characteristic.

[0055] Beneficial effects of the present invention:

[0056] The present invention can achieve a refined assessment of the national land space by dividing the target area into grids and collecting and analyzing different characteristic data. For example, for soil characteristics, a variety of chemical, physical and biological property indicators are comprehensively considered, making the assessment of land fertility more accurate and comprehensive.

[0057] The present invention, an innovative feature analysis method, can effectively evaluate the degree of dispersion of feature data in the target area by calculating the mean value, standard deviation, etc. This helps to discover abnormal values ​​and distribution patterns of the data, and provides a more reliable basis for subsequent planning decisions. For example, in the analysis of soil fertility, the uniformity of fertility distribution in the region can be accurately judged.

[0058] The present invention can provide clear guidance for national land space planning based on the comparison and judgment between the characteristic index and the pre-set spatial planning area index threshold. Planners can clearly understand which functional areas the target area is suitable for or not suitable for, which improves the scientificity and rationality of the planning. For example, it can accurately determine whether a certain area is suitable as an agricultural development area or an ecological functional area.

[0059] The present invention considers the comprehensive influence of multiple feature data and calculates the comprehensive evaluation index, which makes the evaluation of the target area more comprehensive and avoids the limitation of the evaluation of a single feature data.

[0060] The present invention provides a variety of feature data selection and processing methods, such as soil characteristics, terrain characteristics, hydrological characteristics, etc., which can adapt to the diversity of different regions and different planning needs.

[0061] The present invention displays the spatial planning areas recommended by planning and avoided by planning to relevant personnel through information display, thereby improving the transparency and operability of planning, facilitating the participation and supervision of the planning process by all parties, and promoting the rational use and sustainable development of national space. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The present invention will be further described below in conjunction with the accompanying drawings.

[0063] Figure 1It is a flowchart of the land space planning method based on deep learning of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention. Embodiment 1

[0065] See also Figure 1 As shown, the present invention is a land space planning method based on deep learning, comprising the following steps:

[0066] Step 1: Target Division

[0067] Divide the target area into a grid and obtain several unit areas and corresponding grid nodes;

[0068] In this embodiment, the shapes and sizes of the various unit regions are the same;

[0069] Step 2: Data Collection

[0070] At the grid nodes corresponding to the target area, feature data of the corresponding positions are collected;

[0071] Feature data includes:

[0072] Soil characteristics, such as soil fertility, which is divided into several fertility levels based on the results of soil fertility measurements and each fertility level is assigned a predetermined score;

[0073] Among them, soil fertility is calculated based on the quantitative values ​​of multiple indicator data and the preset weight coefficients corresponding to the indicator data:

[0074] In this embodiment, the indicator data includes but is not limited to:

[0075] Soil chemical properties indicators such as organic matter, total nitrogen, total phosphorus, total potassium, alkaline nitrogen, available phosphorus, available potassium, available calcium, available magnesium, available sulfur, available iron, available manganese, available copper, available zinc, available boron, available molybdenum, pH, electrical conductivity and cation exchange capacity;

[0076] Soil physical properties such as soil moisture content, field water holding capacity, soil texture, bulk density, total porosity, capillary porosity, soil thickness, slope, groundwater depth, soil temperature and water-stable aggregates;

[0077] Soil biological traits indicators such as urease, acid phosphatase, invertase, catalase, microbial biomass carbon, microbial biomass nitrogen, fungi, bacteria, actinomycetes, respiration intensity, metabolic entropy and microbial entropy;

[0078] The formula is: L=Fi×Wi, i=1, 2, 3,...n, where n represents the number of corresponding indicator data when calculating the parameter L value, L represents the soil fertility index, Fi is the quantitative value of the i-th indicator data, and Wi is the weight coefficient of the i-th indicator data.

[0079] When the soil fertility index at the corresponding location is at a certain fertility level, the score assigned to the fertility level is used as the quantitative value of the soil characteristics;

[0080] Step 3: Feature Analysis

[0081] First, the quantized values ​​of the feature data corresponding to all grid nodes are obtained, and the feature index of the target area is analyzed based on the quantized values;

[0082] The specific method is as follows:

[0083] Step D1, taking soil characteristics as an example, obtain the soil fertility index corresponding to each grid node, and use it as the quantitative value of the characteristic data, that is, the quantitative value of the soil characteristics, and then mark it as Tj, j = 1, 2, ... m, m represents the number of grid nodes;

[0084] Step D2, calculate the average value of Tj of all grid nodes in the target area and record it as T0p;

[0085] Step D3, calculate the standard deviation of Tj of all grid nodes in the target area through the standard deviation formula and mark it as Tb;

[0086] Among them, the standard deviation Tb represents the degree of dispersion of the quantized value of the characteristic data in the target area;

[0087] Step D4, compare the standard deviation Tb with the preset discrete threshold Ty:

[0088] If Tb>Ty, it means that the discrete degree of the quantized value of the characteristic data in the target area is large;

[0089] Then, the corresponding Tj values ​​are deleted in descending order according to |Tj-T0p| and the standard deviation Tb of the remaining feature data quantization values ​​is calculated accordingly until Tb≤Ty;

[0090] Then, when Tb≤Ty, the number u of deleted Tj is obtained, and the average value of undeleted Tj is calculated and marked as Tpp;

[0091] Then, the corresponding quantity proportion B is calculated through B= (mu) / m;

[0092] Among them, when comparing: if Tb≤Ty, it means that the discrete degree of the quantized value of the characteristic data in the target area is moderate, then the B value is 1, and Tpp=T0p;

[0093] Step D5, through a specific formula:

[0094] , calculate the characteristic index TZ of the corresponding characteristic data of the target area;

[0095] Among them, α1, α2, and α3 are all preset weight coefficients, μ is the correction factor, and the value of μ is 0.65;

[0096] Step 4: Planning and determination

[0097] According to the characteristic index of the target area, it is compared with the predetermined index thresholds corresponding to several pre-classified spatial planning areas to determine:

[0098] The spatial planning area refers to areas with different pre-planned functions, including but not limited to:

[0099] Agricultural Development Area: an area used primarily for agricultural production and related activities;

[0100] Ecological functional area: an area that undertakes important functions such as maintaining ecological balance and providing ecological services;

[0101] Urban development area: an area planned for urban construction and development;

[0102] Mineral resource development zone: an area rich in various mineral resources and where mining and development are being carried out;

[0103] Water resource protection area: an area where water resources are protected and managed with emphasis;

[0104] Tourist resort: an area whose main function is tourism and leisure;

[0105] Marine functional areas: areas divided according to the natural attributes of the ocean and social needs, such as fishing areas, port and shipping areas, etc.;

[0106] Nature reserve: an area where rare wild animals and plants and special ecosystems are strictly protected.

[0107] The comparison and judgment method is as follows:

[0108] StepK1, obtain the index threshold Zy corresponding to each spatial planning area;

[0109] Step K2, then compare the characteristic index TZ of the target area with each index threshold Zy:

[0110] If TZ ≥ Zy, the spatial planning area is marked as reserved, and the corresponding index difference ZC of the corresponding spatial planning area is calculated by ZC = TZ - Zy;

[0111] If TZ<Zy, the spatial planning area will not be marked as reserved;

[0112] Step K3, obtain all spatial planning areas with reservation marks, and recommend the corresponding spatial planning areas as the planning of the target area, and all spatial planning areas without reservation marks as planning avoidance of the target area;

[0113] Planning recommendation refers to the suggestion to set the target area as a certain spatial planning area, and planning avoidance refers to the suggestion not to set the target area as a certain spatial planning area;

[0114] Step K4, sort the spatial planning areas for planning recommendations and planning avoidances according to the order of index differences from large to small, and generate a display form;

[0115] For example, suppose:

[0116] If the characteristic index of the target area is greater than the index threshold of the corresponding soil characteristics of the agricultural development zone and the ecological function zone, the target area can be divided into the agricultural development zone or the ecological function zone;

[0117] If the characteristic index of the target area is less than the index threshold of the corresponding soil characteristics of the agricultural development zone, it is not recommended to plan the target area as an agricultural development zone;

[0118] Step 5: Information display

[0119] The spatial planning areas corresponding to planning recommendations and planning avoidances in the display form will be displayed to relevant personnel.

[0120] This embodiment divides the target area into grids, making the evaluation more refined and accurately grasping the distribution characteristics of soil fertility in the area; comprehensively considering many soil chemical, physical and biological property indicators to calculate the land fertility index, thereby improving the accuracy and comprehensiveness of the evaluation; detailed feature analysis steps, including calculating the mean value, standard deviation, etc., can deeply understand the discrete degree and overall characteristics of soil fertility in the target area; by comparing and judging with the index threshold of the spatial planning area, a clear direction is provided for planning, which is conducive to the rational planning of different functional areas such as agricultural development areas and ecological functional areas. Embodiment 2

[0121] As the second embodiment of the present invention, when the present application is implemented, compared with the first embodiment, the technical solution of the present embodiment is different from that of the first embodiment only in that:

[0122] In this embodiment, the soil characteristics included in the characteristic data in the first embodiment are replaced by terrain characteristics;

[0123] Topographic features, such as altitude, which are divided into several specific altitude intervals and each altitude interval is assigned a predetermined score;

[0124] When the altitude of the corresponding location is within a certain altitude range, the score assigned to the altitude range is used as the quantitative value of the terrain feature;

[0125] This embodiment replaces soil characteristics with terrain characteristics, enriching the evaluation dimensions and making planning considerations more comprehensive; dividing and assigning scores according to specific altitude intervals can clearly understand the impact of the terrain characteristics of the target area on spatial planning; and it is helpful to make reasonable plans based on terrain characteristics, such as planning urban development zones in areas with suitable altitudes, and avoiding over-development in high-altitude or complex terrain areas. Embodiment 3

[0126] As the third embodiment of the present invention, when the present application is implemented, compared with the first embodiment, the technical solution of the present embodiment is different from that of the first embodiment only in that:

[0127] In this embodiment, the soil characteristics included in the characteristic data in the first embodiment are replaced by hydrological characteristics;

[0128] Hydrological characteristics, such as groundwater level, are based on a predetermined reasonable range of groundwater level, and a number of difference intervals are set according to the difference corresponding to the groundwater level exceeding the reasonable range, and a predetermined score is assigned to each difference interval, wherein the reasonable range is also assigned a corresponding predetermined score;

[0129] When the groundwater level at the corresponding location is within a reasonable range, the score assigned to the reasonable range is used as the quantitative value of the hydrological characteristic;

[0130] When the groundwater level at the corresponding location is not within the reasonable range, first compare the groundwater level with the reasonable range:

[0131] If the groundwater level is lower than the reasonable range, the absolute value of the difference between the groundwater level and the minimum value in the reasonable range is calculated, and then the score assigned to the difference interval corresponding to the absolute value of the difference is used as the quantitative value of the hydrological characteristic;

[0132] If the groundwater level is higher than the reasonable range, the absolute value of the difference between the groundwater level and the maximum value in the reasonable range is calculated, and then the score assigned to the difference interval corresponding to the absolute value of the difference is used as the quantitative value of the hydrological characteristic;

[0133] This embodiment introduces hydrological characteristics to make the factors considered in spatial planning more diverse, which can better protect water resources and achieve rational utilization; the difference between groundwater levels and reasonable ranges is divided into intervals and scores are assigned, accurately quantifying the impact of groundwater levels on planning; it provides a scientific basis for the planning of water-related spatial planning areas such as water resource protection areas, and ensures the security of regional water resources. Embodiment 4

[0134] As the fourth embodiment of the present invention, the present application is different from the first embodiment, the second embodiment and the third embodiment in that: the present embodiment is implemented in combination with the methods of the first embodiment, the second embodiment and the third embodiment;

[0135] This embodiment also adds the step of comprehensive analysis after the third step and before the fourth step in the first embodiment.

[0136] Calculate the comprehensive evaluation index of the target area according to the characteristic indexes corresponding to the multiple characteristic data;

[0137] The calculation formula of the comprehensive evaluation index is:

[0138] ;

[0139] Where Z0 represents the comprehensive evaluation index, TZk represents the characteristic index corresponding to each characteristic data, k=1, 2, ... t, t represents the number of characteristic data, and TZp is the average value of multiple characteristic data TZk;

[0140] Then, according to the fourth step in Example 1, the predetermined index threshold corresponding to the comprehensive evaluation index of the spatial planning area is obtained, and it is compared with the comprehensive evaluation index of the target area for determination.

[0141] This embodiment adds a comprehensive analysis step, comprehensively considers multiple feature data, makes the evaluation result more comprehensive and integrated, and avoids the limitations of a single feature; makes planning decisions based on the comprehensive evaluation index, improves the scientificity and rationality of planning decisions, and achieves the balance and optimization of multiple factors. Embodiment 5

[0142] As the fifth embodiment of the present invention, when the present application is specifically implemented, compared with the first, second, third and fourth embodiments, the technical solution of this embodiment is to combine and implement the solutions of the above-mentioned first, second, third and fourth embodiments.

[0143] This embodiment combines and implements the solutions of multiple embodiments, fully integrating various characteristics such as soil, topography, hydrology, etc., to make the planning more complete and accurate; comprehensively considering various factors, it can maximize the rational allocation of resources and the effective use of space to meet various development needs; it provides a comprehensive and systematic solution for complex national land space planning, and improves the adaptability and sustainability of the planning.

[0144] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0145] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A national land space planning method based on deep learning, characterized in that: The following steps are involved: Step 1: Divide the target area into a grid and obtain several unit areas and corresponding grid nodes; Step 2: At the grid nodes corresponding to the target area, collect the characteristic data of the corresponding location, including soil characteristics, terrain characteristics, and hydrological characteristics: Step 3: Obtain the quantized values ​​of the feature data corresponding to all grid nodes, and analyze the feature index of the target area based on the quantized values; In step 3, the analysis method of the characteristic index corresponding to each characteristic data is the same. The soil characteristics are selected and the analysis of its characteristic index is as follows: Step D1, obtain the soil fertility index corresponding to each grid node, and use it as the quantitative value of the characteristic data, that is, the quantitative value of the soil characteristic, and then mark it as Tj, j = 1, 2, ... m, m represents the number of grid nodes; Step D2, calculate the average value of Tj of all grid nodes in the target area and record it as T0p; Step D3, calculate the standard deviation of Tj of all grid nodes in the target area through the standard deviation formula and mark it as Tb; Among them, the standard deviation Tb represents the degree of dispersion of the quantized value of the characteristic data in the target area; Step D4, compare the standard deviation Tb with the preset discrete threshold Ty: If Tb>Ty, it means that the discrete degree of the quantized value of the characteristic data in the target area is large; Then, the corresponding Tj values ​​are deleted in descending order according to |Tj-T0p| and the standard deviation Tb of the remaining feature data quantization values ​​is calculated accordingly until Tb≤Ty; Then, when Tb≤Ty, the number u of deleted Tj is obtained, and the average value of undeleted Tj is calculated and marked as Tpp; Then, the corresponding quantity proportion B is calculated through B= (mu) / m; Among them, when comparing: if Tb≤Ty, it means that the discrete degree of the quantized value of the characteristic data in the target area is moderate, then the B value is 1, and Tpp=T0p; Step D5, through the formula: , calculate the characteristic index TZ of the corresponding characteristic data of the target area; Among them, α1, α2, and α3 are all preset weight coefficients, μ is the correction factor, and the value of μ is 0.65; Step 4: Calculate the comprehensive evaluation index of the target area based on the characteristic indexes corresponding to the multiple characteristic data; Step 5: Compare and judge the characteristic index of the target area with the predetermined index thresholds corresponding to several pre-classified spatial planning areas, and obtain a display form containing planning recommendations and planning avoidance corresponding spatial planning areas based on the comparison and judgment results: Among them, spatial planning area refers to areas with different pre-planned functions, including agricultural development areas, ecological functional areas, urban development areas, mineral resource development areas, and water resource protection areas. Planning recommendation refers to the suggestion to set the target area as a certain spatial planning area, and planning avoidance refers to the suggestion not to set the target area as a certain spatial planning area.

2. The land space planning method based on deep learning according to claim 1 is characterized in that: The calculation formula of the comprehensive evaluation index is: ; Where Z0 represents the comprehensive evaluation index, TZk represents the characteristic index corresponding to each characteristic data, k=1, 2, ...t, t represents the number of characteristic data, and TZp is the average value of multiple characteristic data TZk.

3. The land space planning method based on deep learning according to claim 2 is characterized in that: The comparison and judgment method is as follows: StepK1, obtain the index threshold Z0y corresponding to the comprehensive evaluation index in each spatial planning area; Step K2, then compare the comprehensive evaluation index Z0 of the target area with each index threshold Z0y: If Z0≥Z0y, the spatial planning area is marked as reserved, and the corresponding comprehensive index difference Z0C of the corresponding spatial planning area is calculated by Z0C=Z0-Z0y; Step K3, obtain all spatial planning areas with reservation marks, and recommend the corresponding spatial planning areas as the planning of the target area, and all spatial planning areas without reservation marks as planning avoidance of the target area; Step K4: According to the index difference from large to small, the spatial planning areas recommended for planning and avoided for planning are classified and sorted, and a display form is generated.

4. The land space planning method based on deep learning according to claim 3 is characterized in that: In Step K2, if Z0<Z0y, the spatial planning area will not be marked as reserved.

5. The land space planning method based on deep learning according to claim 1, characterized in that: Soil characteristics are quantified by soil fertility, which is divided into several fertility levels according to the results of soil fertility measurement, and each fertility level is assigned a predetermined score; When the soil fertility index at the corresponding location is at a certain fertility level, the score assigned to the fertility level is used as the quantitative value of the soil characteristics.

6. The method for land space planning based on deep learning according to claim 5 is characterized in that: in, Soil fertility is calculated based on the quantitative values ​​of multiple indicator data and the preset weight coefficients corresponding to the indicator data: The indicator data include organic matter, pH, soil moisture content, invertase, and respiration intensity; The formula is: L=Fi×Wi, i=1, 2, 3,...n, where n represents the number of corresponding indicator data when calculating the parameter L value, L represents the soil fertility index, Fi is the quantitative value of the i-th indicator data, and Wi is the weight coefficient of the i-th indicator data.

7. The method for land space planning based on deep learning according to claim 1, characterized in that: The terrain features are quantified by altitude, which is divided into several specific altitude intervals, and different altitude intervals are assigned predetermined scores; When the altitude of the corresponding location is within a certain altitude range, the score assigned to the altitude range is used as the quantitative value of the terrain feature.

8. The method for land space planning based on deep learning according to claim 1, characterized in that: The hydrological characteristics are quantitatively valued with groundwater level. According to the predetermined reasonable range of groundwater level, several difference intervals are set according to the difference corresponding to the groundwater level exceeding the reasonable range, and a predetermined score is assigned to each difference interval, among which the reasonable range is also assigned a corresponding predetermined score; When the groundwater level at the corresponding location is within a reasonable range, the score assigned to the reasonable range is used as the quantitative value of the hydrological characteristic.

9. The method for land space planning based on deep learning according to claim 8, characterized in that: When the groundwater level at the corresponding location is not within the reasonable range, first compare the groundwater level with the reasonable range: If the groundwater level is lower than the reasonable range, the absolute value of the difference between the groundwater level and the minimum value in the reasonable range is calculated, and then the score assigned to the difference interval corresponding to the absolute value of the difference is used as the quantitative value of the hydrological characteristic; If the groundwater level is higher than a reasonable range, the absolute value of the difference between the groundwater level and the maximum value in the reasonable range is calculated, and then the score assigned to the difference interval corresponding to the absolute value of the difference is used as the quantitative value of the hydrological characteristic.

Images