Urban migration number prediction method

Abstract

The invention discloses an urban migration population prediction method, which belongs to the technical field of geographic space data application, is used for predicting urban migration population, and comprises the following steps: acquiring migration departure place information, migration destination information and migration population; generating grids for the migration departure place and the migration destination; constructing an urban migration number prediction model, and solving a model coefficient and a space damping coefficient; calculating an infrastructure score, a grid environment score and a grid industry investment score after the destination development plan is implemented, and calculating the travel cost from the grid center point of the departure place to the grid center point of the destination; and predicting the number of people migrated from the departure place grid to the destination grid, and summing to obtain the number of people migrated from the departure place grid. Compared with the prior art, the method comprehensively considers the influence of attribute changes of places such as infrastructure, environment quality and industrial investment on crowd migration after development planning implementation, and the prediction result can provide scientific guidance for urban future population flow prediction, public facility layout and livable quality improvement.

Description

Technical Field

[0001] The invention discloses a method for predicting the number of urban migrants, and belongs to the technical field of geographic space data application. Background Art

[0002] Migration refers to the spatial movement of population between two regions, which usually involves a change in residence from the departure point to the destination.

[0003] Many domestic scholars have used gravity models to predict inter-city migration, achieving considerable success. However, these studies often use urban population or GDP as a parameter for prediction, resulting in total migration figures between two cities and lacking a more granular spatial analysis. Furthermore, these studies fail to consider the impact of factors such as living conditions and changes in infrastructure, environmental quality, and industrial investment following the implementation of recent construction plans on population migration, resulting in certain limitations. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for predicting the number of urban migrants to solve the problems in the prior art, such as single prediction parameters, lack of fine-grained spatial analysis, and large limitations in subsequent applications.

[0005] A method for predicting the number of urban migrants, comprising:

[0006] S1. Obtain information on the origin and destination of migration and the number of migrants;

[0007] S2. Generate 1 km x 1 km grids for the migration origin and destination, collect grid indicators, calculate travel distances, and estimate infrastructure scores, grid environmental quality scores, and grid industrial investment scores;

[0008] S3. Build a model to predict the number of urban migrants and solve the model coefficients and spatial damping coefficients;

[0009] S4. Based on the development plan of the migration destination, calculate the infrastructure score, grid environment score, and grid industry investment score after the destination development plan is implemented, and calculate the travel cost from the grid center point of the departure point to the grid center point of the destination;

[0010] S5. Substitute the model coefficients obtained in S3 and the scores obtained in S4 into the urban migrant population prediction model to predict the number of migrants from the departure grid to the destination grid, and sum them up to obtain the number of migrants in the departure grid.

[0011] The migration departure information in S1 is demographic information, including the number of permanent residents and the proportion of high-income population. The migration destination information includes the number of permanent residents in the destination, as well as the attributes of infrastructure, environmental quality, and industrial investment sites. The number of migrants is the number of people whose permanent residence changes from the departure place to the destination within one year.

[0012] The number of permanent residents, high-income population, and migrant population are gridded high-precision population data, among which the migrant population data records the departure and destination of migration.

[0013] Infrastructure includes primary schools, middle schools, hospitals, restaurants, shopping centers, and cultural and entertainment facilities, extracted from the POI basic geographic data source; environmental quality includes vegetation cover and permeable surface, obtained by inverting the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Water Index (NDWI) from the Landsat 8 image data source; industrial investment refers to the funds at the investment destination, obtained from the national industrial and commercial enterprise database data source.

[0014] S2 includes: counting the total permanent population of the grid of migration departure and the proportion of high-income population in the grid, calculating the total permanent population of the grid of migration destination, and estimating the infrastructure score, grid environmental quality score and grid industrial investment score.

[0015] S2 includes: counting the number of people migrating from the departure grid to the destination grid, and calculating the travel cost from the center point of the departure grid to the center point of the destination grid.

[0016] The travel cost from the center point of the departure grid to the center point of the destination grid is measured by distance and obtained based on the road network data using the Dijkstra algorithm.

[0017] S3 includes: substituting the index values ​​of each grid into the model to solve the model coefficient and the spatial damping coefficient. The urban migration population prediction model is:

[0018] T jk =K(S k Z k T k +C k )P j B j exp(-βr);

[0019] Where, T jk represents the number of migrants between the departure grid j and the destination grid k, c k represents the living comfort level of the destination, S k represents the infrastructure score of the destination grid k, Z k represents the environmental quality score of the destination grid k, T k represents the industry investment score of destination grid k, P jrepresents the permanent population size of the departure grid j, B j represents the proportion of high-income population in the departure grid j, r represents the travel cost between the departure grid and the destination grid, β represents the non-negative spatial damping coefficient, and K is an empirical parameter;

[0020] The urban migration population prediction model is an improvement of the gravity model, which is:

[0021] T jk =KP j P k d -βr ;

[0022] Where d represents the distance between the migration origin and the migration destination.

[0023] The infrastructure score, grid environment score, and grid industry investment score in S2 and S4 are standardized scores, with the highest score being 2 and the lowest being 1. The living suitability in the urban migrant population prediction model is a standardized negative score, with the highest score being 0 and the lowest being -1.

[0024] The urban migrant population prediction model is improved by weighting, and the improved urban migrant population prediction model is:

[0025] T jk =K(W S S k Z k T k +W C C k )P j B j exp(-βr);

[0026] Where W S It's S k Z k T k The weight of an item, W C It is C k The weight of

[0027]

[0028] Where, e S It's S k Z k T k The information entropy of an item, e C It is C k Information entropy;

[0029]

[0030] Compared with the existing technology, the present invention has the following beneficial effects: it comprehensively considers the impact of changes in site attributes such as infrastructure, environmental quality, industrial investment, etc. after the implementation of the development plan on population migration, and predicts the total population migrating to each destination grid; the prediction results can provide scientific guidance for future population flow prediction, public facilities layout, and improvement of livability in the city. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flow chart of the method for predicting the number of urban migrants provided by the present invention;

[0032] Figure 2 This is a flow chart of the analysis of the method for predicting the number of urban migrants provided by the present invention;

[0033] Figure 3 This is a distribution map of destination primary schools in an embodiment of the invention;

[0034] Figure 4 This is a distribution map of destination middle schools in an embodiment of the invention;

[0035] Figure 5 This is a distribution map of destination hospitals in an embodiment of the invention;

[0036] Figure 6 This is a destination dining distribution map in an embodiment of the invention;

[0037] Figure 7 This is a destination shopping distribution map in an embodiment of the invention;

[0038] Figure 8 It is a destination culture and entertainment distribution map in the embodiment of the invention;

[0039] Figure 9 is a vegetation coverage distribution map of a destination in an embodiment of the invention;

[0040] Figure 10 It is a distribution map of permeable surfaces at the destination in the embodiment of the invention;

[0041] Figure 11 It is a distribution diagram of investment funds at the destination in the embodiment of the invention;

[0042] Figure 12 This is a distribution map of the degree of living suitability of a destination in an embodiment of the invention. DETAILED DESCRIPTION

[0043] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention are described clearly and completely below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0044] The embodiment of the present invention selects the experimental migration departure point and the experimental migration destination. The flowchart of the urban migration population prediction method provided by the present invention is as follows: Figure 1 As shown, the analysis flow chart of the urban migration number prediction method provided by the present invention is as follows Figure 2 As shown, the experimental steps are:

[0045] Obtain information on the departure and destination of migration and the number of migrants. The information on the departure place of migration includes two demographic characteristics: permanent population and high-income population. The information on the destination of migration includes the number of permanent residents at the destination, as well as three place attributes: infrastructure, environmental quality, and industrial investment. The number of migrants refers to the number of people whose permanent residence changes from the departure place to the destination within one year.

[0046] S11. Obtain high-precision population data on the resident population and high-income population at the experimental migration departure point, the resident population at the experimental migration destination, and the number of migrants in 2021 at a 50m x 50m grid at the experimental migration departure point and experimental migration destination. The migrant population data records both the migration departure point and the destination.

[0047] S12. Obtain POI data for the experimental migration destinations and extract the distribution of six types of POIs: primary schools, middle schools, hospitals, restaurants, shopping, and cultural and entertainment. Obtain Landsat 8 images of the experimental migration destinations and calculate the NDVI (Normalized Difference Vegetation Index) and NDWI (Normalized Difference Water Index) indices using a raster calculator. Extract the funds invested in the experimental migration destinations from the national industrial and commercial enterprise database.

[0048] Generate 1 km × 1 km grids for the departure and destination of migration, count the total permanent population of the grid of the departure place and the proportion of high-income population in the grid, calculate the total permanent population of the destination, measure the grid infrastructure score, grid environmental quality score and grid industrial investment score, count the number of migrants from the departure grid to the destination grid, and calculate the travel cost from the center point of the departure grid to the center point of the destination grid.

[0049] S21. Generate a 1 km × 1 km grid for the experimental migration departure and destination;

[0050] S22. Overlay analysis, summary statistics, and extraction of the permanent population with the experimental migration destination grid to generate the total permanent population of the grid. Overlay analysis, summary statistics, and field calculation of the high-income population with the grid to generate the proportion of the high-income population in the grid.

[0051] S23. Perform a neighbor analysis on the six POIs of primary school, middle school, hospital, restaurant, shopping, and cultural entertainment with the grid center point, calculate the closest distance between the grid center point and each POI, count the maximum and minimum values ​​of the closest distance, and perform a linear transformation so that the highest score is 2 and the lowest score is 1. The distribution map of primary school, middle school, hospital, restaurant, shopping, and cultural entertainment at the migration destination in the embodiment is as follows: Figure 3 、 Figure 4 、 Figure 5 、 Figure 6 、 Figure 7 、 Figure 8 As shown;

[0052] S24. Use the grid to perform regional statistics on NDVI and NDWI to generate a grid mean, calculate the maximum and minimum values ​​of the grid mean, and perform a linear transformation so that the highest score is 2 and the lowest score is 1. The vegetation coverage and permeable surface distribution map of the migration destination are as follows: Figure 9 、 Figure 10 As shown;

[0053] S25. The enterprise investment funds are superimposed on the grid and the total grid investment funds are generated by summarizing the total grid investment funds. The maximum and minimum values ​​of the total grid investment funds are statistically transformed and linearly transformed so that the highest score is 2 and the lowest score is 1. The distribution map of investment funds in the migration destination is shown in the following example. Figure 11 As shown;

[0054] S26. Based on the data from the third national land survey, the residential land area of ​​the grid and the number of permanent residents per unit of residential land in the experimental migration destination are counted, and the residential land area of ​​the grid is multiplied by the number of permanent residents per unit of residential land in the experimental migration destination to obtain the most suitable number of people for living in the grid. If the total number of permanent residents in the grid is consistent with the most suitable number of people for living in the grid, the living suitability score is the highest value (0). Otherwise, the maximum number of excesses and the maximum number of deficiencies of the permanent residents in the grid that exceed or fall below the suitable number of people for living are counted, and linear transformation is performed respectively (the lowest value is -1). The distribution diagram of the living suitability of the migration destination in the embodiment is shown as follows: Figure 12 As shown;

[0055] S27. Perform overlay analysis, summary statistics, and extraction of the number of migrants in 2021 with the departure grid and the destination grid to obtain the number of migrants from the departure grid to the destination grid;

[0056] S28. Based on the Dijkstra algorithm, use the ArcGIS software network analysis tool to calculate the travel distance between the grid center point of each departure point and the grid center point of the destination.

[0057] Based on the above information, a model for predicting the number of urban migrants is constructed, and the index values ​​of each grid are brought into the model to solve for each coefficient. The model for predicting the number of urban migrants is:

[0058] T jk =K(S k Z k T k +C k )P j B j exp(-βr);

[0059] Among them, T jk represents the number of migrants between the departure grid j and the destination grid k, C k represents the living comfort level of the destination, S k represents the infrastructure score of the destination grid k, Z k represents the environmental quality score of the destination grid k, T k represents the industry investment score of destination grid k, P j represents the permanent population size of the departure grid j, B j represents the proportion of high-income population in the departure grid j, r represents the travel cost between the departure grid and the destination grid, β represents the non-negative spatial damping coefficient, and K is an empirical parameter;

[0060] C k Defined as:

[0061]

[0062] Among them, D k is the total permanent population of the k grid, A k is the optimal number of people living in the k-grid.

[0063] The urban migration population prediction model is an improvement of the traditional gravity model, which is:

[0064] T jk =KP j P k d -βr ;

[0065] The infrastructure score, grid environment score, and grid industry investment score in S2 and S4 are standardized scores, with the highest score being 2 and the lowest being 1. The living suitability in the urban migrant population prediction model is a standardized negative score, with the highest score being 0 and the lowest being -1.

[0066] The urban migrant population prediction model is improved by weighting, and the improved urban migrant population prediction model is:

[0067] T jk =K(W S S k Z k T k +WC C k )P j B j exp(-βr);

[0068] Where W S It's S k Z k T k The weight of an item, W C It is C k The weight of

[0069]

[0070] Where, e S It's S k Z k T k The information entropy of an item, e C It is C k Information entropy;

[0071]

[0072] S31. Based on the grid number identifier and the urban migrant population prediction model, generate a matrix of the total number of migrants and the permanent population of the grid, the proportion of high-income population in the grid, the grid's living suitability, the grid's infrastructure score, the grid's environmental quality score, the grid's industrial investment score, and the travel cost;

[0073] S32. Using the regression analysis method, regression analysis was performed and the β was obtained to be 0.00358 and K to be 0.73248.

[0074] According to the development plan of the migration destination, the grid infrastructure score, grid environment score, grid industry investment score and travel cost from the grid center point of the departure point to the grid center point of the destination are calculated after the implementation of the destination development plan.

[0075] S41. Extract the primary, secondary, and hospital facilities in the development plan and merge them with the current facilities. Calculate the closest distance between the grid center and the primary, secondary, and hospital facilities. Count the maximum and minimum values ​​of the closest distances and perform a linear transformation to achieve a maximum score of 2 and a minimum score of 1.

[0076] S42. Extract the content related to vegetation cover and permeable surface in the development plan, adjust the NDVI and NDWI grid means, calculate the maximum and minimum values ​​of the grid means, and perform a linear transformation so that the highest score is 2 and the lowest score is 1;

[0077] S43. Extract the content related to industrial investment in the development plan, adjust the total amount of grid investment funds, calculate the maximum and minimum total amount of grid investment funds, and perform linear transformation so that the highest score is 2 and the lowest score is 1;

[0078] S44. Extract the planned roads in the development plan, merge the planned roads with the existing roads, and use the ArcGIS software network analysis tool based on the Dijkstra algorithm to calculate the travel costs of each departure grid center point and destination grid center point.

[0079] The obtained coefficients, scores and distances are brought into the urban migration population prediction model to predict the number of migrants from the departure grid to the destination grid, and the total population of each grid at the departure point is obtained by summing them up.

[0080] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents, and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the number of urban migrants, characterized in that: include: S1. Obtain information on the origin and destination of migration and the number of migrants; S2. Generate 1 km x 1 km grids for the migration origin and destination, collect grid indicators, calculate travel distances, and estimate infrastructure scores, grid environmental quality scores, and grid industrial investment scores; S3. Build a model to predict the number of urban migrants and solve the model coefficients and spatial damping coefficients; S4. Based on the development plan of the migration destination, calculate the infrastructure score, grid environment score, and grid industry investment score after the destination development plan is implemented, and calculate the travel cost from the grid center point of the departure point to the grid center point of the destination; S5. Substitute the model coefficients obtained in S3 and the scores obtained in S4 into the urban migrant population prediction model to predict the number of migrants from the departure grid to the destination grid, and sum them up to obtain the number of migrants in the departure grid.

2. The method for predicting the number of urban migrants according to claim 1, characterized in that: The migration departure information in S1 is demographic information, including the number of permanent residents and the proportion of high-income population. The migration destination information includes the number of permanent residents in the destination, as well as the attributes of infrastructure, environmental quality, and industrial investment sites. The number of migrants is the number of people whose permanent residence changes from the departure place to the destination within one year.

3. The method for predicting the number of urban migrants according to claim 2, wherein: The number of permanent residents, high-income population, and migrant population are gridded high-precision population data, among which the migrant population data records the departure and destination of migration.

4. The method for predicting the number of urban migrants according to claim 3, characterized in that: Infrastructure includes primary schools, middle schools, hospitals, restaurants, shopping centers, and cultural and entertainment facilities, extracted from the POI basic geographic data source; environmental quality includes vegetation cover and permeable surface, obtained by inverting the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Water Index (NDWI) from the Landsat 8 image data source; industrial investment refers to the funds at the investment destination, obtained from the national industrial and commercial enterprise database data source.

5. A method for predicting the number of urban migrants according to claim 4, characterized in that: S2 includes: Count the total permanent population of the grid at the migration departure point and the proportion of high-income population in the grid, calculate the total permanent population of the grid at the migration destination, and estimate the infrastructure score, grid environmental quality score and grid industrial investment score.

6. A method for predicting the number of urban migrants according to claim 5, characterized in that: S2 includes: Count the number of people migrating from the departure grid to the destination grid, and calculate the travel cost from the center point of the departure grid to the center point of the destination grid.

7. A method for predicting the number of urban migrants according to claim 6, characterized in that: The travel cost from the center point of the departure grid to the center point of the destination grid is measured by distance and obtained based on the road network data using the Dijkstra algorithm.

8. A method for predicting the number of urban migrants according to claim 7, characterized in that: S3 includes: Substitute the index values ​​of each grid into the model to solve the model coefficient and spatial damping coefficient. The urban migration population prediction model is: T jk =K(S k Z k T k +C k )P j B j exp(-βr); Where, T jk represents the number of migrants between the departure grid j and the destination grid k, C k represents the living comfort level of the destination, S k represents the infrastructure score of the destination grid k, Z k represents the environmental quality score of the destination grid k, T k represents the industry investment score of destination grid k, P j represents the permanent population size of the departure grid j, B j represents the proportion of high-income population in the departure grid j, r represents the travel cost between the departure grid and the destination grid, β represents the non-negative spatial damping coefficient, and K is an empirical parameter; The urban migration population prediction model is an improvement of the gravity model, which is: T jk =KP j P k d -βr ; Where d represents the distance between the migration origin and the migration destination.

9. The method for predicting the number of urban migrants according to claim 8, characterized in that: The infrastructure score, grid environment score, and grid industry investment score in S2 and S4 are standardized scores, with the highest score being 2 and the lowest being 1. The living suitability in the urban migrant population prediction model is a standardized negative score, with the highest score being 0 and the lowest being -1.

10. The method for predicting the number of urban migrants according to claim 9, characterized in that: The urban migrant population prediction model is improved by weighting, and the improved urban migrant population prediction model is: T jk =K(W S S k Z k T k +W C C k )P j B j exp(-βr); Where W S It's S k Z k T k The weight of an item, W C It is C k The weight of Where, e S It's S k Z k T k The information entropy of an item, e C It is C k Information entropy;

Images