A land use change simulation prediction method, device, equipment and medium

Abstract

This disclosure relates to a method, apparatus, equipment, and medium for simulating and predicting land use change. The method includes: collecting geographic information point-of-interest data and mobile signaling big data to extract land use data; extracting base year data from a land simulation variable model as driving factors; inputting the land use data and the driving factors into a logistic regression model and calibrating the influence coefficients of each driving factor; extracting dynamic prediction year data from the land simulation variable model, substituting it into the logistic regression model, and calculating to obtain the land simulation prediction result. The method of this disclosure has high accuracy and achieves more detailed land type classification compared to existing Logistic-CA-based land use simulation technology, thus resulting in higher type accuracy. Compared to traditional algorithms, the new technology achieves dynamic simulation by employing spatiotemporal big data and dynamic simulation technology (dynamic iteration and dynamic parameter adjustment), thus achieving higher accuracy on the time scale.

Description

Technical Field

[0001] This application relates to the field of predictive analysis technology, and more specifically, this disclosure relates to a method, apparatus, equipment and medium for simulating and predicting land use change. Background Technology

[0002] Logistic regression, also known as logistic regression analysis, is a generalized linear regression model. The equation for logistic regression is:

[0003] In the formula, β represents the probability of a specific land use type i appearing in each grid cell, β is the regression coefficient of the independent variable, and X represents the data of the selected driving factor. Using logistic stepwise regression to diagnose the probability of a specific land use type appearing in each grid cell can screen out factors that have a significant impact on land cover patterns and determine the quantitative relationships and relative magnitudes of their effects. Cellular automata (CA) is a grid-based statistical dynamic model that simulates the spatiotemporal evolution of complex phenomena based on transformation rules. By employing the ability of cellular automata (CA) to simulate the spatial changes of complex systems and alter the state of cells, dynamic simulation and prediction of land use patterns can be performed.

[0004] Traditional urban land use simulation relies primarily on static data, which can only reflect a specific moment and has poor spatial accuracy. For example, population variable data largely depends on population grid data from the Chinese Academy of Sciences, which is based on traditional population survey data and updated every 5 years. Moreover, due to the high cost of gridding, it only achieves a 1km grid scale, resulting in insufficient spatial accuracy.

[0005] Current technology does not achieve dynamic simulation; the land use probability of each grid is set to be constant, making the algorithm logically unreasonable. Using 2000 land use data and variable data, a spatial distribution probability suitability map of each land use type in 2000 (the probability of a certain land type appearing in each grid cell) is derived through a Logistic regression equation. Then, when predicting land use distribution in 2015, the 2000 land use distribution suitability map is directly used as the 2015 land use distribution suitability map and input into the model. However, in reality, the variable data of each grid is constantly changing. For example, the population in 2000 and the population in 2015 differ greatly. This method is merely a simplification of the actual situation and is logically flawed.

[0006] Traditional land use simulation classification is rather crude and unsuitable for urban land use simulation. Traditional land use simulation uses the LUCC classification, which divides land into several categories such as arable land, forest land, grassland, water area, urban and rural settlements, industrial and mining areas, and unused land. This classification cannot simulate the evolution of construction land within cities.

[0007] The variable system and constraints of traditional land use simulation are insufficient to meet the needs of urban land simulation. Theoretically, traditional land use simulations rarely consider variables within the city itself. For example, due to the imbalance between jobs and residence, the impact of the employed population on urban land use is as significant as that of the resident population. Furthermore, the rise of TOD (Transit-Oriented Development) theory highlights the significant impact of subway stations on land use changes, necessitating the addition of new variables. Moreover, each type of urban land use is subject to constraints such as slope, planning control lines, and engineering conditions, which are rarely addressed in traditional land use models. Summary of the Invention

[0008] To address the key technical bottlenecks and problems of existing urban land use change simulation technologies, such as poor real-time dynamism, unreasonable algorithm logic, outdated variable system, and low spatial accuracy, this disclosure provides a convenient, high-precision method for simulating and predicting urban land use change based on multi-source spatiotemporal big data, which meets the needs of refined urban management.

[0009] To achieve the above technical objectives, this disclosure provides a simulation prediction method, including:

[0010] Collect mobile phone signaling big data and geographic information point of interest data, and extract land use data;

[0011] Base year data were extracted from the land simulation variable model as driving factors;

[0012] Input the land use data and the driving factors into the logistic regression model and calibrate the influence coefficient of each driving factor;

[0013] The dynamic prediction year data extracted from the land simulation variable model are substituted into the logistic regression model, and the land simulation prediction results are obtained through calculation.

[0014] The land simulation prediction results will be displayed graphically or pushed to the user in the form of a message.

[0015] Furthermore, the land use data specifically includes: residential land, commercial and service facilities land, public management and public service land, public utility land, green space and square land, road and transportation facilities land, industrial land and / or non-construction land.

[0016] Furthermore, before calculating and obtaining the land simulation prediction results, the method further includes:

[0017] The land simulation prediction results are corrected using land simulation constraints.

[0018] Furthermore, the constraints for the land simulation are as follows:

[0019] When the slope When the threshold value is reached, the probability of developing the land into the corresponding land type is 0; otherwise, it is 1.

[0020] ;

[0021] In the formula, These are the critical slope values ​​for different land uses.

[0022] Furthermore, the step of extracting dynamic prediction year data from the land simulation variable model, substituting it into the logistic regression model, and calculating the land simulation prediction results specifically includes:

[0023] The study area was divided into 250m grid units. The frequency density and category ratio of geographic information points of interest data for each type of land use in each grid unit were statistically calculated, and the land use type with the largest category ratio was selected as the dominant type of urban land use.

[0024] The calculation formula is:

[0025] ;

[0026] Among them, frequency density It represents the proportion of the i-th type of geographic interest point data within a grid cell to the total number of geographic interest points of that type; ni is the number of i-th type geographic interest points within a grid cell. It represents the total number of geographic interest points of type i in the study area; the category ratio. It is the frequency density of the i-th type of geographic information point of interest data within the grid cell. The proportion of the frequency density of all categories of geographic information points of interest data within this unit;

[0027] At the same time, remote sensing satellite data is used to identify non-construction land. Grids with a non-construction land ratio of more than 80% are identified as non-construction land and overlaid on the land identified by POI to form the final urban land use map.

[0028] The overall probability value of a certain grid property transforming into a certain land type. for:

[0029] ;

[0030] The local conversion probability, i.e., the probability that urban land use unit i is converted to other types of land use at time t, is calculated using the following formula:

[0031] ;

[0032] In the formula, , among which, γ0, γ1,..., γ n These are the weighting coefficients. , , ..., These are the various driving factors of land use conversion;

[0033] For neighborhood interaction, the calculation formula is:

[0034] ;

[0035] It is a conditional function. If the state of a neighboring cell outside the central cell is v, it is assigned a value of 1; otherwise, it is assigned a value of 0. The summation afterward represents the number of neighboring cells with state v.

[0036] This is the land use conversion coefficient, representing the inherent difficulty of converting a certain land type to other land use types;

[0037] The inertia coefficient represents the inheritance of land use type k in the current cell at iteration t, indicating that the land use type remains k. and These represent the differences between the macro-demand and actual allocation of land use type k in the (t-1)th and (t-2)th iterations, respectively. The initial inertia coefficient is set to 1 for both iterations. The calculation formula is as follows:

[0038] .

[0039] Furthermore, the mobile signaling big data specifically includes:

[0040] Information such as age tags, gender tags, residency information, residential grid number, and number of residents in the grid are used to identify and categorize permanent residents, population distribution, and / or population migration data.

[0041] To achieve the above-mentioned technical objectives, this disclosure also provides a simulation prediction apparatus, comprising:

[0042] The data acquisition module is used to collect mobile phone signaling big data and geographic information point of interest data and extract land use data;

[0043] The data extraction module is used to extract base year data from the land simulation variable model as driving factors;

[0044] The calibration module is used to input the land use data and the driving factors into the logistic regression model and calibrate the influence coefficients of each driving factor;

[0045] The calculation module is used to extract the dynamic prediction year data from the land simulation variable model, substitute it into the logistic regression model, and perform calculations to obtain the land simulation prediction results.

[0046] Furthermore, it also includes:

[0047] The correction module is used to correct the land simulation prediction results using land simulation constraints.

[0048] To achieve the above-mentioned technical objectives, this disclosure also provides a computer storage medium storing a computer program thereon, which, when executed by a processor, is used to implement the steps of the above-described simulation prediction method.

[0049] To achieve the above-mentioned technical objectives, this disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described simulation prediction method.

[0050] The beneficial effects of this disclosure are as follows:

[0051] Urban land use simulation and prediction technology based on multi-source dynamic big data shows good predictive performance. Taking Beijing's land use distribution in 2020 as an example, by incorporating updates to dynamic big data and land use conversion coefficients, the simulation can better predict land use changes. Based on comparison with real data, the overall accuracy of the land simulation is 88.16%, indicating good predictive performance.

[0052] Advantages of this disclosure:

[0053] The model boasts high accuracy, achieving more detailed land type classification compared to existing Logistic-CA-based land use simulation techniques, thus resulting in higher type precision. Compared to traditional algorithms, the new technology implements dynamic simulation, employing spatiotemporal big data and dynamic simulation techniques (dynamic iteration and dynamic parameter adjustment), thus achieving higher accuracy on the time scale.

[0054] Economical and practical, this spatiotemporal big data approach boasts a wide scope, high time window and spatial accuracy, and strong real-time performance. The other data used are generally available, low-cost, and easy to collect. In particular, the use of POI data and mobile signaling big data to derive urban land use overcomes the shortcomings of traditional remote sensing satellite data, which suffers from overly coarse classification. Furthermore, it addresses the issues of high cost, long update times, and high confidentiality associated with government land survey data. This technology lowers the barrier to entry for urban land data, enabling more researchers to access high-precision urban land use data.

[0055] The model has high generalizability. Data is readily available, and parameter estimation methods are convenient; the model is simple to operate, highly automated, and easy to promote. Attached Figure Description

[0056] Figure 1 A flowchart illustrating the method of Embodiment 1 of this disclosure is shown;

[0057] Figure 2 This diagram illustrates the accuracy of land use prediction zoning in Beijing in 2020.

[0058] Figure 3 A schematic diagram of the device according to Embodiment 2 of this disclosure is shown;

[0059] Figure 4 A schematic diagram of the structure of Embodiment 4 of this disclosure is shown. Detailed Implementation

[0060] Embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0061] The accompanying drawings illustrate various structural schematics according to embodiments of the present disclosure. These drawings are not to scale, and some details have been enlarged and may have been omitted for clarity. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0062] Example 1:

[0063] like Figure 1 As shown:

[0064] This disclosure provides a simulation prediction method, including:

[0065] Collect mobile phone signaling big data and geographic information point of interest data, and extract land use data;

[0066] Base year data were extracted from the land simulation variable model as driving factors;

[0067] Input the land use data and the driving factors into the logistic regression model and calibrate the influence coefficient of each driving factor;

[0068] The dynamic prediction year data extracted from the land simulation variable model are substituted into the logistic regression model, and the land simulation prediction results are obtained through calculation.

[0069] The land simulation prediction results will be displayed graphically or pushed to the user in the form of a message.

[0070] Furthermore, the land use data specifically includes: residential land, commercial and service facilities land, public management and public service land, public utility land, green space and square land, road and transportation facilities land, industrial land and / or non-construction land.

[0071] Furthermore, before calculating and obtaining the land simulation prediction results, the method further includes:

[0072] The land simulation prediction results are corrected using land simulation constraints.

[0073] Furthermore, the constraints for the land simulation are as follows:

[0074] When the slope When the threshold value is reached, the probability of developing the land into the corresponding land type is 0; otherwise, it is 1.

[0075] ;

[0076] In the formula, These are the critical slope values ​​for different land uses.

[0077] Furthermore, the step of extracting dynamic prediction year data from the land simulation variable model, substituting it into the logistic regression model, and calculating the land simulation prediction results specifically includes:

[0078] The study area was divided into 250m grid units. The frequency density and category ratio of geographic information points of interest data for each type of land use in each grid unit were statistically calculated, and the land use type with the largest category ratio was selected as the dominant type of urban land use.

[0079] The calculation formula is:

[0080] ;

[0081] Among them, frequency density It represents the proportion of the i-th type of geographic interest point data within a grid cell to the total number of geographic interest points of that type; ni is the number of i-th type geographic interest points within a grid cell. It represents the total number of geographic interest points of type i in the study area; the category ratio. It is the frequency density of the i-th type of geographic information point of interest data within the grid cell. The proportion of the frequency density of all categories of geographic information points of interest data within this unit;

[0082] At the same time, remote sensing satellite data is used to identify non-construction land. Grids with a non-construction land ratio of more than 80% are identified as non-construction land and overlaid on the land identified by POI to form the final urban land use map.

[0083] The overall probability value of a certain grid property transforming into a certain land type. for:

[0084] ;

[0085] The local conversion probability, i.e., the probability that urban land use unit i is converted to other types of land use at time t, is calculated using the following formula:

[0086] ;

[0087] In the formula, , among which, γ0, γ1,..., γ n These are the weighting coefficients. , , ..., These are the various driving factors of land use conversion;

[0088] For neighborhood interaction, the calculation formula is:

[0089] ;

[0090] It is a conditional function. If the state of a neighboring cell outside the central cell is v, it is assigned a value of 1; otherwise, it is assigned a value of 0. The summation afterward represents the number of neighboring cells with state v.

[0091] This is the land use conversion coefficient, representing the inherent difficulty of converting a certain land type to other land use types;

[0092] The inertia coefficient represents the inheritance of land use type k in the current cell at iteration t, indicating that the land use type remains k. and These represent the differences between the macro-demand and actual allocation of land use type k in the (t-1)th and (t-2)th iterations, respectively. The initial inertia coefficient is set to 1 for both iterations. The calculation formula is as follows:

[0093] .

[0094] Furthermore, the mobile signaling big data specifically includes:

[0095] Information such as age tags, gender tags, residency information, residential grid number, and number of residents in the grid are used to identify and categorize permanent residents, population distribution, and / or population migration data.

[0096] The research roadmap for urban land use simulation and prediction technology based on multi-source dynamic big data is mainly divided into the following five parts:

[0097] (1) This technology first subdivides urban land into eight types according to the purpose and nature of the development project: residential land, commercial service facilities land, public management and public service land, public facilities land, green space and square land, road and transportation facilities land, industrial land, and non-construction land.

[0098] (2) Land function identification using POI big data and mobile signaling big data. The categories of POI data are mapped to land classifications. A weighted approach is used to assign POI saliency based on public perception, spatial distribution, and individual characteristics. Weights are assigned to each type of POI facility, and the frequency density of each type of POI in each grid unit is calculated. The POI with the highest frequency density is selected as the dominant land use function of the grid. Then, mobile signaling big data is used to correct the accuracy of land function identification. The quantitative relationship between residents' travel activities and land use functions is calculated using artificial intelligence machine learning. High-precision identification of residents' travel activities is performed using mobile signaling big data, thereby correcting the land functions identified from the POI data.

[0099] (3) Construct an algorithm for urban land use change model. By integrating urban evolution theory, TOD theory and Alonso rent theory, an algorithm for urban land use change model is constructed to perform synchronous system calculations on variables such as population, employment, housing prices, transportation accessibility, and subway stations (see the following for the specific algorithm).

[0100] Table 1 Simulation Variables of Urban Land Use

[0101]

[0102] (4) Dynamically update the development probability of each land use in each grid of the city. Using land use data and variable data from the base year (2015), the coefficients of each variable are obtained through the Logistic regression equation. Then, when predicting the land use distribution in the target year (2020), the corresponding variable data for the target year are dynamically updated to calculate the land use distribution probability in the target year (2020), instead of directly using the land use distribution probability of the base year (2015) as the land use distribution suitability atlas for the prediction year (2020) and inputting it into the model, as in the traditional method. Finally, the probability of each of the eight land uses appearing in each grid is calculated.

[0103] (5) Land use simulation and prediction are carried out by incorporating constraints such as topography, land use characteristics, and urban spatial control policies. First, according to existing regulations, topographic slope restricts land development: the slope of residential land is less than 25%; the slope of industrial land is less than 10%; and the slope of road land is less than 5%. If the slope in the grid exceeds this range, the conversion probability of the corresponding land type is set to 0. Second, due to the differences in land characteristics, the probability of converting one type of land use to another is different. There are a total of 8×8 land conversion coefficients for the 8 types of land use. The land conversion coefficients are used to correct the land conversion probability. Then, urban land use changes are constrained and affected by the red line for cultivated land and the red line for ecological protection. Therefore, the simulated land use results need to be adjusted. Finally, the cellular automata iterates continuously according to the input land use demand until the quantity of each type of land use meets the demand, thus determining the final land use type for each grid.

[0104] The method and technical process scheme disclosed herein are as follows: Figure 1 As shown.

[0105] (1) Data preparation:

[0106] ① Mobile signaling big data:

[0107] Mobile phone signaling is the communication record data between a mobile phone and a communication base station. When a mobile phone connects to a mobile communication network, it generates a series of control commands. The data fields of these commands include various information such as time, location, and number. In this disclosure, mobile phone signaling data is mainly used to obtain urban categorized permanent resident population distribution data and urban categorized permanent resident population residence and migration data. Specifically, it includes information such as age tags (minors, working-age population, elderly population), gender tags (male, female), residency information, residential grid number, and the number of residents in the grid to identify categorized permanent residents, population residence distribution, and population migration.

[0108] ②Basic geographic information data:

[0109] Basic geographic information refers to the most universal and widely shared basic geographic units used by almost all industries related to geographic information for unified spatial positioning and spatial analysis. It mainly consists of elements from natural geographic information such as landforms, water systems, and vegetation, as well as elements from social geographic information such as settlements, transportation, boundaries, special features, and place names. In this disclosure, administrative division data of Beijing is primarily used as the base map for spatial geographic information, and data from subways, transportation networks, and bus stops are used to create variable data for land simulation.

[0110] ③POI data:

[0111] POI is an abbreviation for "Point of Interest." In Geographic Information Systems (GIS), a POI can be a building, a shop, a mailbox, a bus stop, etc., and each POI contains four pieces of information: name, category, coordinates, and classification. In this disclosure, POI data is mainly used to simulate urban land use distribution.

[0112] (2) Calculation method:

[0113] Urban land use inversion based on POI data;

[0114] This method assigns weights to POI salience based on three characteristics: public perception, spatial distribution, and individual features. The specific weight values ​​are as follows:

[0115] Table 2 Correspondence between Urban Land Types and POI Categories

[0116]

[0117] The study area was divided into 250m grid cells. The frequency density and category proportion of POIs for each land use type within each grid cell were statistically calculated. The land use type with the highest category proportion was selected as the dominant urban land use type. The calculation formula is as follows:

[0118] ;

[0119] Among them, frequency density It represents the proportion of the i-th type of geographic interest point data within a grid cell to the total number of geographic interest points of that type; ni is the number of i-th type geographic interest points within a grid cell. It represents the total number of geographic interest points of type i in the study area; the category ratio. It is the frequency density of the i-th type of geographic information point of interest data within the grid cell. The proportion of the frequency density of all categories of geographic information points of interest data within this unit;

[0120] Because POI points are relatively scarce in suburban areas, there is a certain degree of error in identifying non-construction land. Therefore, remote sensing satellite data is used to distinguish between urban construction land and non-construction land. Grids where the proportion of non-construction land identified by satellite reaches more than 80% are identified as non-construction land and overlaid on the land identified by POIs to form the final urban land use map.

[0121] Then, mobile signaling big data is used to correct the accuracy of land use identification. Artificial intelligence and machine learning are applied to construct a quantitative relationship between residents' travel activities and land use functions. Based on this quantitative relationship, mobile signaling big data is used to accurately identify residents' travel activities, and the land use functions identified from POI data are corrected.

[0122] Calculate the conversion probability of each land use in each grid.

[0123] The overall probability value of a certain grid property transforming into a certain land type. for:

[0124] ;

[0125] The local conversion probability, i.e., the probability that urban land use unit i is converted to other types of land use at time t, is calculated using the following formula:

[0126] ;

[0127] In the formula, , among which, γ0, γ1,..., γ n These are the weighting coefficients. , , ..., These are the various driving factors of land use conversion;

[0128] For neighborhood interaction, the calculation formula is:

[0129] ;

[0130] It is a conditional function. If the state of a neighboring cell outside the central cell is v, it is assigned a value of 1; otherwise, it is assigned a value of 0. The summation afterward represents the number of neighboring cells with state v.

[0131] This is the land use conversion coefficient, representing the inherent difficulty of converting a certain land type to other land use types;

[0132] The inertia coefficient represents the inheritance of land use type k in the current cell at iteration t, indicating that the land use type remains k. and These represent the differences between the macro-demand and actual allocation of land use type k in the (t-1)th and (t-2)th iterations, respectively. The initial inertia coefficient is set to 1 for both iterations. The calculation formula is as follows:

[0133] .

[0134] Determine the constraints of the model:

[0135] When the slope When the threshold value is reached, the probability of developing the land into the corresponding land type is 0; otherwise, it is 1.

[0136] ;

[0137] In the formula, These are the critical slope values ​​for different land uses.

[0138] Because urban land use changes are constrained and influenced by the red lines for arable land and ecological protection, the final simulated land use results need to be adjusted to restrict the occupation of non-construction land within the red lines. Land use in other areas will be allocated according to the results of a roulette wheel. The specific calculation formula is as follows:

[0139] ;

[0140] In the formula, This refers to the land use allocation results after adjustments based on policy constraints.

[0141] Below is a specific example of the method disclosed herein:

[0142] Based on multi-source dynamic big data, the urban land use simulation predicts the land use distribution in Beijing in 2020. The specific details are as follows:

[0143] Data Acquisition:

[0144] The specific data sources for the 2020 Beijing land distribution forecast are shown in Table 1. They mainly include mobile phone signaling data, basic geographic information data, statistical census data, POI data, and housing price data.

[0145] Table 3. Data sources and brief information used

[0146]

[0147] (2) Establishing a model

[0148] First, urban land use distribution is inverted based on POI data. Second, the parameters of the variables are calibrated by regression using a logistic model based on land use data from the base year (2015) and variables such as population, employment, housing prices, and transportation. Then, based on the variable data of the prediction year (2020), the local conversion probability of each grid in the prediction year (2020) is obtained. At the same time, the overall probability is corrected according to the slope constraint, each land conversion coefficient, and the red line constraint. The land use transformation process of each cell is simulated by cellular automata to obtain the land use distribution map of the prediction year (2020). Finally, the land function is corrected using mobile signaling big data.

[0149] (3) Prediction results

[0150] In terms of prediction accuracy, the predicted results of land use in Beijing in 2020 are in good agreement with the actual results, with an overall accuracy of 88.16%.

[0151] Residential land is distributed within the city's Sixth Ring Road and in the central areas of outlying districts (Changping, Yanqing, etc.), coinciding with the distribution of residential communities. Commercial service facilities land is relatively concentrated, coinciding with Beijing's major commercial districts (Xidan, Wangfujing, Wangjing, etc.) and surrounding residential land. Public management and public service land, as well as public utility land, are distributed similarly to commercial service land. Green spaces and plazas are mainly distributed in the Tiananmen area of ​​the city's core, near the Botanical Garden, Fragrant Hills, and Olympic Forest Park within the Fifth Ring Road. Industrial land is mainly located in industrial parks outside the Fifth Ring Road, such as the Beijing Economic-Technological Development Area.

[0152] Based on the urban area, near-suburbs, and far-suburbs of Beijing, the accuracy of land use forecasting was statistically calculated by region. The urban area includes Dongcheng District, Xicheng District, Chaoyang District, Haidian District, Fengtai District, and Shijingshan District; the six near-suburbs include Daxing District, Tongzhou District, Shunyi District, Changping District, Mentougou District, and Fangshan District; and the four far-suburbs include Huairou District, Pinggu District, Miyun District, and Yanqing District. According to the regional accuracy calculation results, the overall accuracy of the central urban area is 60.23%, the near-suburbs are 85.40%, and the far-suburbs are 94.92%, all showing high accuracy.

[0153] Example 2:

[0154] like Figure 3 As shown:

[0155] This disclosure also provides a simulation prediction device, comprising:

[0156] Data acquisition module 201 is used to collect mobile phone signaling big data and geographic information point of interest data and extract land use data;

[0157] Data extraction module 202 is used to extract base year data from the land simulation variable model as driving factors;

[0158] The calibration module 203 is used to input the land use data and the driving factors into the logistic regression model and calibrate the influence coefficients of each driving factor;

[0159] The calculation module 204 is used to extract the dynamic prediction year data from the land simulation variable model, substitute it into the logistic regression model, and perform calculations to obtain the land simulation prediction results.

[0160] The push module 205 is used to push the land simulation prediction results to the user in the form of a graphic display or a message.

[0161] The data acquisition module 201 described in this disclosure is sequentially connected to the data extraction module 202, the calibration module 203, the calculation module 204, and the push module 205.

[0162] Furthermore, it also includes:

[0163] The correction module is used to correct the land simulation prediction results using land simulation constraints.

[0164] Example 3:

[0165] This disclosure also provides a computer storage medium storing a computer program thereon, which, when executed by a processor, is used to implement the steps of the above-described simulation prediction method.

[0166] The computer storage medium disclosed herein can be implemented using semiconductor memory, magnetic core memory, magnetic drum memory, or disk memory.

[0167] Semiconductor memory, primarily used in computers, mainly consists of two types of semiconductor storage elements: MOSFETs and bipolar transistors. MOSFETs offer high integration density and simple manufacturing processes but are relatively slow. Bipolar transistors have complex manufacturing processes, high power consumption, and low integration density but are fast. The advent of NMOS and CMOS technologies led to MOSFETs becoming the dominant semiconductor memory. NMOS is fast; for example, Intel's 1K-bit static random access memory (SRAM) has an access time of 45ns. CMOS, on the other hand, consumes less power; a 4K-bit CMOS SRAM has an access time of 300ns. The semiconductor memories mentioned above are all random access memories (RAM), meaning they can be randomly read from and written to during operation. Semiconductor read-only memories (ROMs), however, can be randomly read from but not written to during operation; they are used to store pre-programmed programs and data. ROMs are further divided into non-rewritable fuse-type read-only memories (PROMs) and rewritable read-only memories (EPROMs).

[0168] Magnetic core memory is characterized by low cost and high reliability, and has over 20 years of practical application experience. Before the mid-1970s, magnetic core memory was widely used as main memory. Its storage capacity could reach 10 bits or more, with the fastest access time being 300 ns. Typical international magnetic core memory capacities ranged from 4 MS to 8 MB, with access cycles of 1.0 to 1.5 μs. Even after the rapid development of semiconductor memory replaced magnetic core memory as the main memory, magnetic core memory can still be used as a large-capacity expansion memory.

[0169] Magnetic drum memory is a type of external storage device that records magnetic data. Due to its fast data access speed and stable, reliable operation, although its capacity is relatively small and it is gradually being replaced by disk storage, it is still used as external storage for real-time process control computers and medium- to large-scale computers. To meet the needs of small and microcomputers, ultra-miniature magnetic drums have emerged, which are small in size, lightweight, highly reliable, and easy to use.

[0170] Disk storage is a type of external storage device that records magnetic data. It combines the advantages of magnetic drums and magnetic tapes: its storage capacity is larger than that of magnetic drums, its access speed is faster than that of magnetic tapes, and it can be stored offline. Therefore, disks are widely used as high-capacity external storage in various computer systems. Disks are generally divided into two main categories: hard disks and floppy disks.

[0171] There are many types of hard disk storage devices. Structurally, they are divided into two types: replaceable and fixed. Replaceable disks have interchangeable platters, while fixed disks have fixed platters. Both replaceable and fixed disks have multi-platter and single-platter structures, and can be further divided into fixed-head and movable-head types. Fixed-head disks have smaller capacities, lower recording densities, and higher access speeds, but are more expensive. Movable-head disks have higher recording densities (up to 1000-6250 bits / inch), resulting in larger capacities, but their access speeds are relatively lower than fixed-head disks. Disk products can have storage capacities of several hundred megabytes, with a bit density of 6250 bits / inch and a track density of 475 tracks / inch. Multi-platter replaceable disk storage devices, due to their replaceable platters, offer very large independent capacity, high speed, and are widely used in online information retrieval systems and database management systems for storing large amounts of information.

[0172] Example 4:

[0173] This disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described simulation prediction method.

[0174] Figure 4 This is a schematic diagram of the internal structure of an electronic device in one embodiment. For example... Figure 4As shown, the electronic device includes a processor, a storage medium, a memory, and a network interface connected via a system bus. The storage medium stores an operating system, a database, and computer-readable instructions. The database may store a sequence of control information. When executed by the processor, the computer-readable instructions enable the processor to implement a simulation prediction method. The processor provides computational and control capabilities, supporting the operation of the entire computer device. The memory stores computer-readable instructions, which, when executed by the processor, enable the processor to perform a simulation prediction method. The network interface of the computer device is used for communication with a terminal. Those skilled in the art will understand that… Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0175] This electronic device includes, but is not limited to, smartphones, computers, tablets, wearable smart devices, artificial intelligence devices, and power banks.

[0176] In some embodiments, the processor may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It performs various functions and processes data by running or executing programs or modules stored in the memory (e.g., executing remote data read / write programs) and calling data stored in the memory.

[0177] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory and at least one processor, etc.

[0178] Figure 4 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 4The structure shown does not constitute a limitation on the electronic device and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0179] For example, although not shown, the electronic device may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor via a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be elaborated further here.

[0180] Furthermore, the electronic device may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device and other electronic devices.

[0181] Optionally, the electronic device may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device and to display a visual user interface.

[0182] Furthermore, the computer's usable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store the operating system, applications required for at least one function, etc.; and the data storage area may store data created based on the use of blockchain nodes, etc.

[0183] In the several embodiments provided in this disclosure, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0184] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0185] Furthermore, the functional modules in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0186] This disclosure is of the following significance:

[0187] Technological Innovation:

[0188] The innovative application of spatiotemporal big data in urban land use simulation has been publicly demonstrated. While methods for characterizing population flow based on mobile phone signaling data have gained increasing attention, few studies have incorporated this data into the variable system of urban land use simulation. Urban population distribution and employment distribution have a significant impact on urban land layout, but previously, data availability issues hindered their application in urban land use simulation. The use of spatiotemporal big data allows for the inclusion of population and employment in the variable system, resulting in a more detailed and realistic depiction of urban land use evolution. The dynamic nature and fine spatiotemporal granularity of spatiotemporal big data (grid accuracy down to 250m) also make dynamic, high-precision land use simulation possible.

[0189] A publicly developed "Dynamic Simulation Model of Urban Land Use with Spatio-temporal Big Data" (DSMUL) has been constructed. This technology utilizes big data such as mobile phone signaling and Points of Interest (POIs) to achieve dynamic simulation through algorithm design, establishing a real-time link between driving factors and land use changes. It also creatively incorporates population, employment, housing prices, subway stations, and transportation behavior into the variable system, further aligning with urban development theories and laws, and strengthening the theoretical support for urban land use simulation. Furthermore, considering the impact of the complex urban environment on urban land use evolution, slope constraints, spatial control constraints, and the difficulty coefficient of transformation within construction land are incorporated into the model, further enriching the model's logical framework, fully estimating the complexity of the land use evolution process, and facilitating a more comprehensive and in-depth understanding of urban land use evolution patterns, thereby improving the model's simulation accuracy.

[0190] Subdividing urban land use types to suit urban land use simulations is crucial. Previous simulation studies of urban land use evolution have largely treated urban land as a single land use type (construction land), distinguishing it from non-construction land. However, in reality, urban land use types include residential land, industrial land, commercial and service land, and public facilities land, etc., and the driving factors and related stakeholders for the expansion of different types of land use are different. Therefore, each land use type and the conversion between them needs to be treated specifically. Refining urban land use types is fundamental to further exploring the complex patterns of urban land use conversion.

[0191] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A land use change simulation and prediction method based on spatiotemporal big data, characterized in that, include: Collect geographic information point of interest data and mobile signaling big data, and extract land use data; Base year data were extracted from the land simulation variable model as driving factors; Input the land use data and the driving factors into the logistic regression model and calibrate the influence coefficient of each driving factor; The dynamic prediction year data extracted from the land simulation variable model are substituted into the logistic regression model, and the land simulation prediction results are obtained through calculation. The land simulation prediction results will be displayed graphically or pushed to the user in the form of a message; The process of extracting dynamic prediction year data from the land simulation variable model, substituting it into the logistic regression model, and calculating the land simulation prediction results specifically includes: The study area was divided into 250m grid units. The frequency density and category ratio of geographic information points of interest data for each type of land use in each grid unit were statistically calculated, and the land use type with the largest category ratio was selected as the dominant type of urban land use. The calculation formula is: ； Among them, frequency density It represents the proportion of the i-th type of geographic interest point data within a grid cell to the total number of geographic interest points of that type; ni is the number of i-th type geographic interest points within a grid cell. It represents the total number of geographic interest points of type i in the study area; the category ratio. It is the frequency density of the i-th type of geographic information point of interest data within the grid cell. The proportion of the frequency density of all categories of geographic information points of interest data within this unit; The overall probability value of a certain grid property transforming into a certain land type. for: ； in, For local transition probabilities, That is, the probability that urban land use unit i is converted to other types of land use at time t, calculated by the following formula: ； Where, , among them, γ0, γ1,…, γ n These are the weighting coefficients. , , ..., These are the various driving factors of land use conversion; For neighborhood interaction, the calculation formula is: ； in, It is a conditional function. If the state of a neighboring cell outside the central cell is v, it is assigned a value of 1; otherwise, it is assigned a value of 0. The summation afterward represents the number of neighboring cells with state v. This is the land use conversion coefficient, representing the inherent difficulty of converting a certain land type to other land use types; The inertia coefficient, This indicates that when the land use type of the current cell is k, it will still be k in the t-th iteration. and Let denot represent the difference between the macro-level demand and the actual allocation quantity of land use type k in the (t-1)th and (t-2)th iterations. The initial inertia coefficient is set to 1 for both iterations, and the calculation formula is: ； The mobile signaling big data specifically includes: Age tags, gender tags, residency information, residential grid number, and number of residents in the grid.

2. The method according to claim 1, characterized in that, The land use data specifically includes: residential land, commercial and service facilities land, public management and public service land, public utility land, green space and square land, road and transportation facilities land, industrial land and / or non-construction land.

3. The method according to claim 1, characterized in that, After obtaining the land simulation prediction results through calculation, the method further includes: The land simulation prediction results are corrected using land simulation constraints.

4. The method according to claim 3, characterized in that The constraints for the land simulation are as follows: When the slope When the threshold value is reached, the probability of developing the land into the corresponding land type is 0; otherwise, it is 1. ； In the formula, These are the critical slope values ​​for different land uses.

5. A simulation prediction device, characterized in that, include: The data acquisition module is used to collect mobile phone signaling big data and geographic information point of interest data and extract land use data; The data extraction module is used to extract base year data from the land simulation variable model as driving factors; The calibration module is used to input the land use data and the driving factors into the logistic regression model and calibrate the influence coefficients of each driving factor; The calculation module is used to extract the dynamic prediction year data from the land simulation variable model, substitute it into the logistic regression model, and perform calculations to obtain the land simulation prediction results. The push module is used to display the land simulation prediction results graphically or push them to the user in the form of a message; The process of extracting dynamic prediction year data from the land simulation variable model, substituting it into the logistic regression model, and calculating the land simulation prediction results specifically includes: The study area was divided into 250m grid units. The frequency density and category ratio of geographic information points of interest data for each type of land use in each grid unit were statistically calculated, and the land use type with the largest category ratio was selected as the dominant type of urban land use. The calculation formula is: ； Among them, frequency density It represents the proportion of the i-th type of geographic interest point data within a grid cell to the total number of geographic interest points of that type; ni is the number of i-th type geographic interest points within a grid cell. It represents the total number of geographic interest points of type i in the study area; the category ratio. It is the frequency density of the i-th type of geographic information point of interest data within the grid cell. The proportion of the frequency density of all categories of geographic information points of interest data within this unit; The overall probability value of a certain grid property transforming into a certain land type. for: ； in, For local transition probabilities, That is, the probability that urban land use unit i is converted to other types of land use at time t, calculated by the following formula: ； Where, , among them, γ0, γ1,…, γ n These are the weighting coefficients. , , ..., These are the various driving factors of land use conversion; For neighborhood interaction, the calculation formula is: ； in, It is a conditional function. If the state of a neighboring cell outside the central cell is v, it is assigned a value of 1; otherwise, it is assigned a value of 0. The summation afterward represents the number of neighboring cells with state v. This is the land use conversion coefficient, representing the inherent difficulty of converting a certain land type to other land use types; The inertia coefficient, This indicates that when the land use type of the current cell is k, it will still be k in the t-th iteration. and Let denot represent the difference between the macro-level demand and the actual allocation quantity of land use type k in the (t-1)th and (t-2)th iterations. The initial inertia coefficient is set to 1 for both iterations, and the calculation formula is: ； The mobile signaling big data specifically includes: Age tags, gender tags, residency information, residential grid number, and number of residents in the grid.

6. The apparatus according to claim 5, characterized in that, Also includes: The correction module is used to correct the land simulation prediction results using land simulation constraints.

7. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps corresponding to the simulation prediction method described in any one of claims 1 to 4.

8. A computer storage medium storing computer program instructions thereon, characterized in that, When the program instructions are executed by the processor, they are used to implement the steps corresponding to the simulation prediction method described in any one of claims 1 to 4.

Images