Urban resource enjoyment unevenness dynamic evaluation method based on reachable time constraint

Abstract

The invention provides an urban resource enjoyment unevenness dynamic evaluation method based on reachable time constraint, which comprises the following steps: independently constructing a supply end model, a demand end model and a connection end model, when any parameter of a supply end, a demand end or a connection end changes, only updating data of an affected model, and when any parameter of the supply end, the demand end or the connection end changes, updating the data of the affected model; recalculating an unevenness evaluation result based on the dynamic reachability; according to a dynamic interaction relationship among the supply end capability, the demand end weight and the connection end accessibility, generating a space resource enjoyment unevenness quantitative index; and in combination with the unevenness quantitative index, a new resource site selection scheme is generated through a particle swarm algorithm, and the scheme meets the minimum spacing constraint, the service coverage rate constraint and the functional diversity constraint.

Description

Technical Field

[0001] The present invention belongs to the technical field of spatial resource fairness assessment and dynamic optimization in the field of smart cities, and specifically relates to a dynamic assessment method for inequality in urban resource enjoyment based on reachable time constraints, and the cross-application of reachable time modeling, real-time path planning, and multi-objective optimization algorithms based on multi-source data fusion (remote sensing, population, and transportation). Background Art

[0002] Current assessments of urban resource spatial inequality are mainly based on accessibility analysis methods, including:

[0003] Buffer analysis and minimum neighbor distance methods: They measure accessibility using Euclidean distance, ignoring the impact of the actual traffic network on travel costs;

[0004] Network analysis method: Calculates path distance based on the traffic network, but relies on high-precision road network data and has high computational complexity;

[0005] Two-step moving search method (2SFCA): This method combines the supply-demand relationship with the distance decay function to evaluate reachability, but it has the following limitations:

[0006] The dichotomy method used to handle distance decay (e.g., a fixed threshold to divide travel intention into reachable and unreachable) does not reflect the continuous decay of travel intention with distance.

[0007] Setting a single fixed search threshold does not consider the impact of different resource sizes and travel modes on service scope;

[0008] Path calculation is mostly based on static Euclidean distance or simplified road network, ignoring the actual path time differences of multiple entry resource points and multiple transportation modes.

[0009] The above method faces the following key problems in practical application:

[0010] Static modeling limitations: It cannot dynamically respond to changes in urban resource supply (such as new facilities), demand (such as population migration), and connectivity (such as road network optimization), resulting in delayed assessment results.

[0011] Distance decay bias: Fixed thresholds and dichotomy processing are difficult to accurately simulate residents' actual travel behavior (e.g., short-distance sensitivity and rapid decay over medium and long distances);

[0012] Path calculation distortion: This ignores the real-time path planning differences among multiple transportation modes (walking, public transportation, and driving), as well as the need to select the optimal path for multiple entry resource points.

[0013] Insufficient scalability: Traditional models couple supply and demand side parameters with connection side parameters. Local data updates require global recalculation, making it difficult to support large-scale urban dynamic assessments. Summary of the Invention

[0014] To address the shortcomings of existing technologies, such as static modeling limitations, distance decay bias, and insufficient dynamic response, this paper provides a dynamic assessment and optimization method for urban resource spatial inequality based on reachable time constraints. This method focuses on the decoupled modeling and dynamic interaction mechanism of the supply side, demand side, and connection side. Specifically, it includes:

[0015] 1. Three-terminal independent modeling and loosely coupled interaction:

[0016] Supply side: Quantify resource supply capacity based on the Remote Sensing Ecological Index (RSEI), and dynamically evaluate supply capacity by integrating multi-dimensional ecological indicators;

[0017] Demand side: Dynamic demand weights are constructed through cellular grid segmentation and population distribution data;

[0018] Connection end: Balance the supply and demand relationship based on reachable time, call the online map API to obtain the real-time path time of multiple travel modes, and calculate dynamic accessibility by combining the supply and demand matching degree (POD) and Gaussian decay function.

[0019] 2. Incremental update mechanism: When new resources are added on the supply side, population distribution changes on the demand side, or road network optimization occurs on the connection side, only local data updates and dynamic reachability recalculations are triggered, breaking through the inefficient bottleneck of traditional global recalculations.

[0020] 3. Multi-constraint site selection optimization: Generate new resource plans based on the particle swarm algorithm, integrating minimum spacing constraints, service coverage, and functional diversity corrections to adapt to the refined planning needs of urban green spaces, medical and educational facilities.

[0021] The present invention solves the problem that traditional methods are unable to respond to dynamic changes in resources through a dynamic interaction framework with three-end decoupling and an incremental update mechanism, providing scalable technical support for urban resource fairness assessment and spatial planning.

[0022] The technical solution specifically adopted by the present invention to solve the technical problem is:

[0023] A dynamic evaluation method for inequality in urban resource access based on accessibility time constraints includes the following steps:

[0024] Independently build the supply-side model, demand-side model, and connection-side model, including:

[0025] Independently build the supply-side model, demand-side model, and connection-side model, including:

[0026] The supply-side model quantifies resource supply capacity through ecological quality indicators and eliminates inconsistencies in quantified resource supply capacity through normalization;

[0027] The demand-side model generates demand distribution weights based on a spatial gridded demand distribution model and population distribution data. In actual use, in order to reduce the number of supply and demand-side nodes, a cellular grid partitioning or other similar methods can be used.

[0028] The connection end model calls the online map API to obtain the real-time shortest path time between the supply and demand ends under multiple travel modes, and dynamically calculates the comprehensive supply service volume enjoyed by each demand end based on the dynamic interactive relationship between the supply and demand matching degree POD and the reachable time TOD;

[0029] When any parameter on the supply side, demand side, or connection side changes, only the data of the affected model is updated, and the inequality assessment results are recalculated based on dynamic accessibility;

[0030] Based on the dynamic interactive relationship between supply-side capacity, demand-side weight and connection-side accessibility, a quantitative indicator of inequality in access to spatial resources is generated.

[0031] Combined with the quantitative indicators of inequality, a new resource location scheme is generated through a particle swarm algorithm, which satisfies the minimum spacing constraint, service coverage constraint and functional diversity constraint.

[0032] Furthermore, the ecological quality indicator is the Remote Sensing Ecological Index (RSEI), which is generated by the following steps:

[0033] Extract greenness, humidity, dryness and heat indexes from remote sensing images, including:

[0034] The greenness index is calculated by the reflectance of the near-infrared band and the red band;

[0035] The humidity index is calculated by the humidity component of the tasseled cap transformation;

[0036] Normalizing the greenness, humidity, dryness and heat indices;

[0037] A comprehensive score is generated based on principal component analysis to quantify resource supply capacity.

[0038] Furthermore, the dynamic interaction relationship is calculated by the following method to obtain dynamic reachability:

[0039] Defining the supply and demand matching POD ij and TOD ij The product of is the comprehensive barrier factor:

[0040] PTOD ij =POD ij ×TOD ij

[0041] According to the dynamically adjusted time threshold T th , calculate the attenuation coefficient Lod through the Gaussian functionij ,satisfy:

[0042]

[0043] Based on the attenuation coefficient, calculate the comprehensive service value of the demand point

[0044] Furthermore, updating the data of the affected model includes the following situations:

[0045] When a new supply point is added, only the supply capacity data of the supply-side model is updated, and the dynamic accessibility with the affected demand points is recalculated;

[0046] When the population distribution on the demand side changes, only the demand weight data of the demand side model is updated, and the reachable time of the connection end is re-matched;

[0047] When the traffic network is optimized, only the path time data of the connection end is updated through the online map API, and the decay function is recalculated.

[0048] Furthermore, the objective function of the particle swarm algorithm is:

[0049]

[0050] Where: F is the total cost; represents the distance cost term, which is used to minimize the sum of the reachable distances; αS is the service coverage reward term, which is used to maximize the service coverage of the newly built park; P and λ represent the penalty term and the corresponding penalty weight, respectively, which are used to deal with the violation of the constraint conditions;

[0051] D ij is the Euclidean distance from demand point i to newly built resource j;

[0052] S is the number of demand points covered by the newly built resources;

[0053] P unreach 、P no_serve 、P new 、P exist These are the penalty items for demand point unreachable, newly built resources not serving, insufficient spacing between newly built resources, and insufficient spacing between newly built resources and existing facilities;

[0054] Constraints include:

[0055] Newly created resources meet the minimum spacing constraint;

[0056] The service coverage rate meets the preset threshold;

[0057] Maintain a minimum safe distance from existing facilities.

[0058] Furthermore, the supply-side modeling further includes: combining the functional mix of surrounding points of interest (POIs) to correct the comprehensive score of resource supply capabilities.

[0059] Furthermore, the calculation of the Euclidean distance Dij includes: performing path planning on multiple entry points of the same resource, and selecting the entry with the minimum time cost as the valid path.

[0060] Furthermore, the calculation of the shortest path time includes: performing path planning on multiple entry points of the same resource, and selecting the entry with the minimum time cost as the valid path.

[0061] And, a dynamic evaluation system for inequality in urban resource access based on reachable time constraints, including:

[0062] Supply-side modeling module: used to quantify resource supply capacity through the Remote Sensing Ecological Index (RSEI);

[0063] Demand-side modeling module: used to generate demand distribution weights based on cellular grid subdivision and population distribution data;

[0064] Connection end modeling module: Calls the online map API to obtain the real-time shortest path time between supply and demand sides under multiple travel modes, and dynamically calculates the comprehensive supply service volume enjoyed by each demand side based on the dynamic interactive relationship between the supply and demand matching degree (POD) and the reachable time (TOD);

[0065] Dynamic incremental update module: When any parameter on the supply side, demand side, or connection side changes, it triggers local data update and recalculates the inequality assessment results;

[0066] Inequality Assessment Module: Generates quantitative indicators of inequality in access to spatial resources based on the dynamic interaction between supply-side capacity, demand-side weight, and connection-side accessibility;

[0067] Multi-constraint optimization module: Generates new resource location plans that meet minimum spacing constraints, service coverage constraints, and functional diversity constraints through particle swarm optimization.

[0068] And, a dynamic evaluation system for implementing the above method, comprising:

[0069] Data input interface, used to access remote sensing images, population distribution and traffic network data;

[0070] a processor configured to perform supply-side modeling, demand-side modeling, connection-side modeling, dynamic incremental updating, inequality assessment, and multi-constraint optimization algorithms;

[0071] Output interface for generating resource optimization site selection plans and inequality visualization reports.

[0072] And, an electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the program.

[0073] A non-transitory computer-readable storage medium stores a computer program, which implements the steps of the method described above when executed by a processor.

[0074] Compared with the prior art, the present invention and its preferred embodiments have at least the following beneficial effects:

[0075] 1. Dynamic modeling and incremental update advantages: Through independent modeling on three sides and a loosely coupled interaction mechanism, it supports local updates when any parameter on the supply side, demand side, or connection side changes independently, avoiding the resource waste of traditional global recalculation and significantly improving evaluation efficiency and flexibility.

[0076] 2. Accurately balance supply and demand imbalances at the connection end: Based on real-time path planning and the Gaussian decay function, this approach dynamically quantifies the regulatory effect of reachable time on the supply and demand relationship, resolving evaluation biases caused by fixed thresholds and static paths in traditional methods (such as 2SFCA).

[0077] 3. Scientifically optimize and adapt to diverse needs: Generate new resource site selection plans through a multi-constraint optimization algorithm, taking into account fairness (minimum spacing constraints), efficiency (service coverage), and urban function adaptability (POI diversity correction), thereby improving the feasibility of planning solutions.

[0078] 4. Universal expansion of the technical framework: Through modular design, it is compatible with the dynamic assessment needs of different types of urban resources (such as green space, medical care, and education), providing a reusable method system for urban space governance. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0080] Figure 1 This is a schematic diagram of the "supply-demand" end structure of the connection end balance according to an embodiment of the present invention.

[0081] Figure 2 Schematic diagram of supply-side spatial modeling for estimating green ecological space supply capacity based on RSEI in an embodiment of the present invention.

[0082] Figure 3 Schematic diagram of demand-side spatial modeling according to an embodiment of the present invention, wherein (a) shows the distribution of residential communities, (b) shows the honeycomb grid division of the community, and (c) shows the aggregated community demand points.

[0083] Figure 4This is a schematic diagram of the spatial modeling of the connection end according to an embodiment of the present invention, where (a), (c), and (e) are the shortest reachable times from the community to the nearest supply point by walking, public transportation, and driving, respectively; (b), (d), and (f) are the average reachable times to the supply point by the three modes; and (g), (h), and (i) are comparison histograms.

[0084] Figure 5 Schematic diagram of the calculation process of the comprehensive spatial service model enjoyed by the demand point in an embodiment of the present invention.

[0085] Figure 6 This is a spatial distribution map of community park and comprehensive park services enjoyed by residents under different travel modes and time thresholds in an embodiment of the present invention.

[0086] Figure 7 This is a K-means clustering curve diagram of an embodiment of the present invention.

[0087] Figure 8 This is a convergence curve diagram of the objective function of an embodiment of the present invention.

[0088] Figure 9 This is a schematic diagram of the newly added park green space in an embodiment of the present invention, where (a) is the spatial distribution of the newly added park green space, and (b) is the distribution space of the ecological environment quality RSEI of the newly added park green space and the surrounding POI services.

[0089] Figure 10 This is a diagram showing the incremental accessibility of park green spaces under different travel modes and time thresholds after a new park is added in an embodiment of the present invention. DETAILED DESCRIPTION

[0090] In order to make the features and advantages of the present invention more clearly understood, the following embodiments are given for detailed description:

[0091] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in this specification have the same meaning as commonly understood by those skilled in the art to which this application belongs.

[0092] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0093] This paper addresses the dynamic assessment of spatial inequality in urban resource access and proposes a technology for balancing supply and demand based on reachable time (connection end). This technology clarifies and designs key structures, including conceptual models, mathematical definitions, and calculation methods for the supply, demand, and connection ends, to address the challenges of dynamic modeling and assessment of spatial inequality in resource access. The research results are expected to be applied to the optimal allocation, site selection, and spatial planning simulation of urban spatial resources, such as the site selection, assessment, and planning of urban green spaces, medical resources, and public service facilities such as education.

[0094] Specifically, the structure of the reachable time (connection end) balancing the "supply-demand" end designed in this embodiment is as follows: Figure 1 As shown in Figure 1, reachable time is the bridge connecting the supply side and the demand side. The premise of connecting the supply and demand relationship through the bridge is that the supply side and the demand side match. Whether the supply and demand sides match determines the necessity of connection. Therefore, it is necessary to reconstruct the three ends of "supply-demand-connection". The supply side mainly considers the location of the supply point. i}、Supply quantity {N oi}、Supply quality {Q oi}, the comprehensive supply capacity of the supply point {S oi} can be achieved through To express. The demand side mainly constructs the demand point location {D j},Required quantity {N Dj},Requirement Quality {Q Dj}, comprehensive demand capacity of demand points {S Dj} can be obtained through S Dj =N Dj ×Q Dj The connection end mainly considers the supply and demand matching (POD) calculation, such as the comprehensive service capability of the supply point. Less than the demand point S Dj (Right now ), connect and S Dj It loses its meaning, so its matching degree POD ij In addition to considering the matching degree, the connection end also needs to calculate the connection time, that is, the time from the demand end to the supply end TOD, which mainly depends on the connection method and its convenience. ij ) Function LOD can be established according to the reachable time decay ij =f(POD,TOD).

[0095] 1. Supply-side (O) modeling

[0096] Combining multi-source data such as remote sensing images, the spatial supply capacity of urban resources can be evaluated. Although different resources may have great differences in service supply quality, scope, direction and measurement methods, they can be expressed by normalizing the comprehensive supply capacity value of the supply point (or expressing it in percentage form). When considering the service capacity of green ecological space, it is usually possible to use a similar Remote Sensing Ecological Index (RSEI) to establish the supply capacity value of the resource at the supply point, such as Figure 2 The RSEI calculation formula is as follows:

[0097] RSEI=f(NDVI,WET,LST,NDBSI) (1)

[0098] Among them, NDVI is the greenness index; WET is the humidity index; NDBSI is the dryness index; LST is the heat index.

[0099] Among them, the greenness NDVI can be calculated by dividing the difference between the reflectance value of the near-infrared band and the reflectance value of the red band in the remote sensing image by the sum of the two. The calculation formula is:

[0100]

[0101] Where p n represents the reflectivity in the near-infrared band; p r Represents the reflectance of the red band.

[0102] Humidity WET can be represented by the humidity component of the Tasseled Cap Transform. This varies between different sensors. For Landsat OLI TIRS images, the calculation formula is:

[0103]

[0104] Where p b 、p g 、p s1 and p s2 They represent the reflectance of the blue band, green band, shortwave infrared band I and shortwave infrared band II of the Landsat image respectively.

[0105] The dryness index (NDSBI) can be composed of the building index (Index-based Built-up Index, IBI) and the bare soil index (Soil Index, SI), as follows:

[0106] NDSBI=(SI+IBI) / 2 (4)

[0107] in,

[0108]

[0109] The heat index can be expressed by the surface temperature, as follows:

[0110]

[0111] Where, LST is the surface temperature; T B is the temperature at the sensor; λ is the center wavelength; ρ is the reflectivity in the band; em is the emissivity.

[0112] In order to avoid the influence of the differences in units and importance of the above four component indicators, principal component analysis is used to extract the first principal component and then normalize the indicators to map the values to the range of [0,1]. The formula is as follows:

[0113] RSEI0=1-PCL1[f(NDVI,WET,LST,NDBSI)] (8)

[0114]

[0115] Among them, RSEI0 is the initial ecological index, PCL1 is the first principal component, and RSEI is the remote sensing ecological index to be determined.

[0116] 2. Demand-side (D) Modeling

[0117] In this embodiment, the demand-side spatial modeling mainly considers the spatial distribution and structural characteristics of demand, such as the demand for green ecological space, which mainly considers the spatial distribution and age structure of the community population. Although it may be difficult to obtain data and other difficulties, it is sometimes impossible to accurately estimate the demand quantity and quality of some demand points (such as the inability to accurately obtain the demand structure of the community population), the spatial distribution of the demand side can still be established according to the distribution of population size, and the spatial structure of the demand side can be updated when new data is collected. Figure 3 The spatial modeling of community population can be carried out using data such as residential areas or night light remote sensing. After dividing the study area into (honeycomb grids), the spatial distribution model of demand points can be established by aggregating residential areas.

[0118] 3. Connection end (L) modeling

[0119] The key to balancing the supply side (O) and the demand side (D) by using the connection side (Lod) in this embodiment is to construct a function that decays and constrains according to the reachable time. Assume

[0120] PTOD ij =PID ij ×TOD ij (10)

[0121] Then the attenuation and constraint functions can be expressed as:

[0122]

[0123] Where, Lod ij Represents supply point O i Able to connect demand point D j The decay rate of the comprehensive service capacity of the supply, corresponding to each independent demand point D j , the comprehensive space service value it can enjoy is:

[0124]

[0125] Through the above method, given different reachable time thresholds T th Under the constraints, the comprehensive service status of the entire spatial demand point can be generated. The actual calculation of the reachable time TOD ij When connecting to a network, you need to set different travel modes (such as walking, public transportation, and driving). Figure 4 It shows the shortest reachable time of each OD under different travel modes, which can be obtained by calling the Baidu or Amap API service.

[0126] IV. Dynamic Assessment of Spatial Inequality

[0127] This embodiment further considers that the evaluation of spatial inequality can actually be transformed into a measure of the level of resources or services enjoyed by different spatial locations. Specifically, in the actual calculation of a demand point D j When the comprehensive service capability value enjoyed by the premises is Figure 5 As shown, first calculate the total supply capacity of different supply points (obtained through the supply-side capacity assessment introduced earlier) Secondly, calculate the TOD from each community point to different supply points i (Under a specific travel mode), the supply and demand matching degree POD between them is superimposed i The PTOD of all communities can be calculated by formula (10): i value, and finally set the time threshold T th Then, the total service capacity value actually obtained by the demand point from all supply points can be calculated according to formula (12).

[0128] According to the above-mentioned design scheme, this embodiment is based on the main urban area of Fuzhou City (within the third ring road, with a total area of about 173.04 km 2 ) Taking park green space resources as an example, dynamic evaluation and application analysis of inequality are carried out.

[0129] (1) Evaluation and analysis of park green space resource supply and service capabilities

[0130] To quantitatively assess the ecological and environmental quality of the parks and green spaces in the study area, this example uses Landsat 8-9OLI / TIRS Collection 2 Level-2 dataset imagery provided by the U.S. Geological Survey (USGS) Earth Resources Observation and Science Center (https: / / earthexplorer.usgs.gov / ). Remote sensing imagery covering the study area from April to October 2022, during the vegetation growing season, was selected. Cloud cover was consistently below 10%, and the spatial resolution was 30 meters. Specific image information is shown in Table 1.

[0131] Table 1 Research remote sensing image data information

[0132]

[0133] The processing of satellite remote sensing data is carried out based on the Google Earth Engine (GEE) cloud platform. The GEE platform provides a large number of free access remote sensing data sets, and users can access and use these data through the Internet API interface and the Web's interactive development environment.

[42] . Its huge data resource reserves and powerful algorithm processing capabilities effectively overcome the problem of cumbersome image data downloading and processing procedures. The image data processing process of this embodiment has the following specific steps: (1) Loading data. Load the vector boundary of the study area and integrate the 2022 Landsat 8 (LANDSAT / LC08 / C02 / T1_L2) and Landsat 9 (LANDSAT / LC09 / C02 / T1_L2) image datasets; (2) Declouding. Use the QA band to write the declouding function code, traverse each image, and remove the interference of cloudy and rainy conditions on the ground object recognition and image classification results; (3) Band selection and correction. According to the research needs, select 7 bands in the image: blue, green, red, near infrared, shortwave infrared 1, shortwave infrared 2 and thermal infrared. The bands are further corrected. A scaling factor is applied to the optical band to convert the DN value into reflectance. The scaling factor and offset information are derived from the metadata of the Landsat image. A scaling factor is also applied to the thermal infrared band to convert the DN value into brightness temperature. The corrected optical and thermal infrared bands are then used to replace the original bands. (4) Cropping. The processed image is cropped according to the vector map of the study area.

[0134] The attractiveness of a park is directly dependent on the quality of its green space—that is, the ecological benefits it provides to residents. This quality of green space influences the frequency and duration of residents' exposure to green space. This study used formulas for four component indices: greenness (NDVI), humidity (Wet), dryness (NDBSI), and heat (LST). These four components were calculated and generated into corresponding layers. Principal component analysis (PCA) was then performed after normalization. Based on the cumulative contribution rate of the PCA, the principal components with the highest contribution rates were retained and transformed. Finally, the RSEI (Resource Specific Index) was used to determine the spatial distribution of the ecological and environmental quality of the park green spaces in the study area.

[0135] (2) Reachable time analysis based on OD cost

[0136] To quantify the spatial convenience of residents reaching parks and green spaces, an initial matrix containing 74,184 pairs of OD combinations was constructed, with the 843 community centroids in the study area as the demand starting points and the 88 entrance points of 43 parks and green spaces as the service end points. The Baidu Map API route planning interface was called to obtain pair-by-pair path planning data for the three travel modes of walking, public transportation, and driving, and the travel time required for residents to reach all park and green space entrances was recorded. For parks and green spaces with multiple entrances, the shortest path screening was also required: for each community centroid, the travel time to all entrances of the same park was calculated, and only the valid path with the minimum time cost was retained. Finally, 36,249 pairs of valid OD time cost data were obtained (Table 2), which only included the travel time from each community centroid to the nearest entrance of the park and green space.

[0137] Table 2 OD time cost of community centroid-park green space travel based on Baidu Map API route planning

[0138]

[0139]

[0140] (3) Evaluation of residents’ access to community parks and comprehensive park green space services under different travel modes and time thresholds

[0141] Figure 6 It reflects the spatial distribution characteristics of residents' accessibility to community parks and comprehensive park services under different travel modes (walking, public transportation, driving) and different time thresholds (15 minutes, 30 minutes).

[0142] (4) Optimizing the location of parks and green spaces

[0143] The site selection process for new urban parks was conducted based on the location entropy matching results of park green areas within community centroids. Specifically, a network of cells where the location entropy remained below 1 at the 30-minute threshold under three travel modes was identified as areas with insufficient park green space resources. The optimal number of park green space locations was determined based on the elbow points of the K-means clustering convergence curve. Subsequently, an objective function was constructed, adhering to the principles of minimizing travel costs and maximizing spatial service efficiency. A particle swarm optimization algorithm was then used to determine the optimal locations for the proposed park green spaces. Finally, the framework was incorporated into the spatial equity assessment of accessibility to analyze the actual effectiveness of the proposed new park green spaces in alleviating unequal access to resources among residents.

[0144] 4.1 Determination of the number of new park green spaces

[0145] For the three travel modes of walking, public transportation and driving, 539 community centroids with location entropy less than 1 within 30 minutes of travel were extracted, and the X and Y coordinates of each point were generated. Subsequently, the K-means clustering algorithm was implemented in the MATLAB environment. Since the kmeans function uses random initial cluster centers by default, the random process is not fixed, resulting in different initial centers each time it is run, and differences in the final clustering results and distance calculations. To solve this problem, the rng function is used to fix the random seed. This embodiment presets the maximum number of clusters to 50 and the seed value to 5. After running the program, a clustering curve is obtained, as shown in Figure 2. Figure 7 shown.

[0146] The horizontal axis of the clustering curve represents the number of cluster centers, and the vertical axis represents the maximum distance from a point to a cluster center. According to the elbow method principle of K-means clustering, the optimal number of clusters corresponds to the inflection point in the curve where the downward trend changes from fast to slow. By observing the clustering curve, it can be found that the inflection point is between 6 and 20 clusters. When the number of clusters exceeds 20, the rate of decrease in distance slows down significantly, and gradually stabilizes in the subsequent period. In this embodiment, 20 is selected as the optimal number of clusters, that is, 20 new park sites are planned to be added.

[0147] 4.2 Determine the objective function and constraints

[0148] After determining the number of parks to be built, the site selection for the new parks can be carried out. In order to ensure that the new parks can meet the needs of the centroid units of each community in the area to be optimized as efficiently as possible, the site selection process needs to comprehensively consider the spatial layout of existing park green spaces and the actual needs of community units for park green spaces, while avoiding the problem of idle resources caused by excessive overlap of park service areas. This embodiment starts from the two main goals of minimizing the distance from the demand point to the nearest new park and maximizing the park green space service coverage, converts the problem of site selection for the proposed parks into a multi-objective optimization problem, and then uses the PSO algorithm to solve the problem.

[0149] This example establishes a comprehensive cost objective function F, the goal of which is to minimize the sum of the distances from residential areas to the nearest newly built park, while maximizing the number of community points served by the park green space, and ensuring that the constraints are met through a penalty term. The objective function is defined as follows:

[0150]

[0151] Where F is the total cost; represents the distance cost term, which is used to minimize the sum of the reachable distances; αS is the service coverage reward term, which is used to maximize the service coverage of the newly built park; P and λ represent the penalty term and the corresponding penalty weight, respectively, which are used to deal with the violation of the constraint conditions.

[0152] ① Distance cost item:

[0153] The sum of the minimum Euclidean distances from all settlement points i to their nearest accessible newly built park needs to be minimized.

[0154] N i ={j∈N|D ij ≤D0}, D0=1000 (14)

[0155] Where M is the set of settlement points i, Create a new park j within the reach of residential area i, D ij is the Euclidean distance between residential point i and newly built park j.

[0156] ②Service coverage reward items:

[0157] The number of covered settlements is multiplied by the reward coefficient α, which is transformed into a minimization objective through the negative sign, that is, maximizing service coverage.

[0158]

[0159] Where S represents the number of settlements served by the newly built park, II(·) is the indicator function, and when settlement i has a newly built park that can be reached, When , the function value is 1, otherwise it is 0; α is the weight coefficient for balancing distance and coverage.

[0160] ③Penalty items:

[0161] Imposing high penalties for violating constraints forces the model to meet planning requirements.

[0162]

[0163] Where, Ⅱ(·) is the indicator function, which has a value of 1 when the conditions in the brackets are met, and 0 otherwise; P unreach As a penalty for residential areas not having access to newly built parks, P no_serveAs a penalty for newly built parks without serving residential areas, P new As a penalty for insufficient spacing between new parks, is the Euclidean distance between the newly built parks j and k, P exist is the penalty for insufficient distance between new and old parks, O is the set of existing parks o; λ represents the corresponding penalty weight, where λ1=λ3=λ4=10 6 denote the penalty weights for inaccessible settlements, insufficient spacing between new parks, and insufficient spacing between old and new parks, respectively. λ2=10 8 The penalty weight for newly built parks without serving settlements.

[0164] The specific requirements of the constraints are:

[0165] (1) Each newly built park must serve at least one residential area. If a newly built park does not serve any residential area, a higher penalty will be added;

[0166] (2) The distance between newly built parks cannot be less than the set minimum distance of 1000m. If the distance between newly built parks is less than the set minimum distance, a penalty item will be added;

[0167] (3) The distance between the new park and the old park cannot be less than the set minimum distance of 1000m. If the distance between the new park and the old park is less than the set minimum distance, a penalty item will be added;

[0168] (4) Each residential area must have at least one accessible park. If a residential area does not have an accessible park, a penalty will be added. 4.3 Optimization of site selection for new parks and green spaces

[0169] This embodiment is based on the defined comprehensive cost objective function and constraints, and constructs a particle swarm optimization algorithm (PSO) solution framework in MATLAB software. Specifically, the 20 parks proposed by the K-means clustering algorithm are used as the initial population of the particle swarm. During the operation of the PSO algorithm, the position of the particles will be continuously updated. At each iteration, the particles will dynamically adjust their coordinates in space based on the individual optimal solution and the group optimal solution, and use the fitness function to calculate the maximum value Gbest of the objective function after each iteration. It continues to iterate and update until the Gbest value meets the algorithm termination condition. It can be considered that the global approximate optimal solution has been found. At this time, the spatial position corresponding to the particle swarm is the output result of the algorithm. The parameter settings of the PSO algorithm are shown in Table 3.

[0170] Table 3 Particle swarm optimization algorithm parameter settings

[0171]

[0172]

[0173] In the data in the table above, the number of particles represents the 20 planned parks and green spaces. The default value of 2 is used for the adjustment weights of individuals and groups. The maximum number of iterations serves as the algorithm's termination threshold. When the number of iterations reaches this value or the change in the optimal value of the objective function, Gbest, falls below the preset tolerance, the algorithm terminates. X and Y represent the horizontal and vertical coordinates of the particle in the spatial coordinate system, respectively. The maximum speeds of X and Y are usually set to 10% to 20% of the variation range. The dimension of the function to be optimized refers to the number of unknowns in the objective function.

[0174] The convergence curve of the objective function Gbest solved by the particle swarm optimization algorithm is as follows: Figure 8 As shown in the figure, the Gbest value decreases rapidly during the first 100 iterations of the algorithm. After 500 iterations, the convergence curve gradually flattens. After 2000 iterations, the algorithm reaches its termination threshold, outputting the X and Y values of 20 particles. At this point, the spatial location formed by the particle swarm is the derived optimal site for the new park green space.

[0175] The coordinates of the proposed park green space location obtained by the particle swarm optimization algorithm are shown in Table 4, where X represents the horizontal coordinate of the center of the proposed park green space, and Y represents the vertical coordinate of the center of the proposed park green space. The specific spatial distribution of the proposed park is shown in Figure 9 (a).

[0176] Table 4 Coordinates of the proposed park green space site

[0177]

[0178] 6.3 Accessibility Equity Analysis Based on New Construction Site Selection

[0179] 6.3.1 Accessibility Spatial Analysis Based on New Construction Site Selection

[0180] In order to achieve the optimal layout of newly built park green spaces, this embodiment uses a combination of K-means clustering algorithm and particle swarm optimization algorithm to optimize the number and spatial location of newly built community parks. Through multi-objective optimization calculation, the spatial distribution of 20 newly added community parks is finally determined as follows: Figure 9 New community parks are primarily located in two types of areas: first, areas with high residential density and limited green space resources. These areas are highly populated, yet underserved by existing park green space, leading to a significant supply-demand imbalance. Second, urban fringe areas are severely lacking in park green space, yet have a high potential demand for such services.

[0181] According to the recommendations on the scale of community parks in the Urban Green Space Planning Standard (GB / T51346-2019), and considering the balance between service capacity and land resource utilization efficiency, the study sets the area of new community parks at 5hm2.2 , to avoid insufficient service capacity due to too small an area, or waste of land resources due to too large an area. 2 Community parks are included in the equity assessment framework proposed in the study, and the extent to which the supply-side incremental renewal plan improves the inequality of park green spaces in the study area is quantitatively analyzed.

[0182] Figure 9 (b) presents the RSEI of the newly built park green space and the functional diversity of its surrounding points of interest (POIs). The study evaluated the comprehensive service provision capacity of newly built community parks based on their size, ecological quality, and surrounding functional diversity. The results are shown in Table 5. The data show that the comprehensive service provision capacity ranges from 0.371 to 0.526, with newly built park green spaces of different numbers varying in their ability to provide comprehensive services to residents. This result provides a data basis for the subsequent calculation of the cumulative service provision capacity available to residents within their accessible area.

[0183] Table 5 RSEI, surrounding functional mix and comprehensive supply service capacity of newly built park green spaces

[0184]

[0185]

[0186] To further quantify the effect of newly added parks in the study area on improving residents' access to park green space supply services, this example reconstructs the OD cost matrix of 20 newly added community parks and 843 community centroids. By calling the route planning interface of the Baidu Map API, path data for three travel modes, namely walking, public transportation, and driving, is obtained, and the travel time required from the community demand point to the newly built park is calculated accordingly. Through the key link of accessibility time, the supply service capacity of park green space is connected with the population demand of the community, and the actual contribution of the site selection of newly added park green space to improving the accessibility of resource services for urban residents is quantitatively evaluated.

[0187] After the addition of parks, the park green space supply service capacity available to community residents under three different travel modes of walking, public transportation, and driving and different time thresholds is improved as follows: Figure 10As shown. Specifically, when walking for 15 minutes, the accessibility increase in most areas is in the range of 0-0.40, and only the local areas around the newly added parks have a more significant increase in accessibility; when the travel time reaches 30 minutes, the degree of improvement in accessibility and the coverage area are significantly increased. For public transportation, it is difficult for residents to achieve effective accessibility in a short period of time, but when it is extended to 30 minutes, the scope of the area with a high degree of improvement is significantly expanded, indicating that public transportation is more ideal for improving access to park and green space services under longer travel times. Regardless of the length of the travel time, the driving mode generally has a higher increase in accessibility, which has a significant advantage in improving residents' access to park and green space services, and the improvement effect becomes more obvious as the travel time increases.

[0188] In summary, the newly added park plan obtained through this embodiment has alleviated the supply-demand imbalance to a certain extent, positively improving the accessibility of park green space to residents across different travel modes and time thresholds, and increasing the convenience of surrounding residents to parks and green spaces. Driving mode has the most significant improvement, while walking and public transportation modes have the most significant improvements for longer trips.

[0189] Considering the differences in spatial supply and demand of resources at different times and locations, accessible time is an effective means to bridge the supply and demand sides and balance such differences. Although some traditional modeling methods have considered the supply and demand imbalance problem, such as two-step mobile search (2SFCA) and its related models, they usually ignore the changes in the supply and demand relationship and the connection end, such as the addition of a certain amount of resource supply, changes in the spatial distribution or demand structure of the demand side, and optimization and upgrading of the connection end, the traditional supply and demand integrated model method is difficult to support dynamic and incremental update problems. The present invention disassembles and analyzes the imbalance relationship between spatial supply and demand, and independently models the spatial supply side, demand side, and connection side of the resources to construct a dynamic evaluation technology for spatial enjoyment inequality under the constraint of accessible time.

[0190] This paper deeply analyzes and decomposes the comprehensive capacity assessment of the supply side of each resource space, the spatial distribution modeling of the demand side, and the reachable time of different travel modes (connection side). Through the loose coupling of the supply side, demand side, and connection side, it constructs a framework and related technologies that support the dynamic update of spatial inequality. The key points to be protected are:

[0191] (1) “Dynamic modeling and incremental updating of spatial inequality”: Construct mathematical models of the spatial supply side (spatial service capacity), demand side (population distribution demand structure) and connection side (OD reachable time) of resources. By disassembling and reconstructing “supply-demand-connection” into a structure of “connection” balancing (alleviating) the spatial imbalance of “supply-demand”, a spatial access inequality assessment framework and method are established to support the quantification of the impact of independent changes in “connectivity”, “supply service capacity” or “demand structure and distribution” on spatial inequality, and to support dynamic modeling and incremental updating of inequality.

[0192] This technology can provide new technologies and methods for related research problems such as resource optimization allocation and spatial planning and site selection. Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors, and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.

[0193] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium having a computer program stored thereon, which executes the above method when executed by a processor. The storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component.

[0194] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0195] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other manner. Any person skilled in the art may utilize the above-disclosed technical content to modify or modify the present invention into equivalent embodiments. However, any simple modifications, equivalent variations, and modifications to the above embodiments that do not depart from the technical content of the present invention and are based on the technical essence of the present invention remain within the scope of protection of the present invention.

[0196] The present invention is not limited to the above-mentioned optimal implementation mode. Anyone can derive various other forms of a dynamic evaluation method for inequality in urban resource access based on reachable time constraints under the inspiration of the present invention. All equivalent changes and modifications made within the scope of the patent application of the present invention should fall within the scope of the present invention.

Claims

1. A dynamic evaluation method for inequality in urban resource access based on reachable time constraints, characterized by: The following steps are involved: Independently build the supply-side model, demand-side model, and connection-side model, including: The supply-side model quantifies resource supply capacity through ecological quality indicators and eliminates inconsistencies in quantified resource supply capacity through normalization; The demand-side model generates demand distribution weights based on the spatial grid demand distribution model and population distribution data; The connection end model calls the online map API to obtain the real-time shortest path time between the supply and demand ends under multiple travel modes, and dynamically calculates the comprehensive supply service volume enjoyed by each demand end based on the dynamic interactive relationship between the supply and demand matching degree POD and the reachable time TOD; When any parameter on the supply side, demand side, or connection side changes, only the data of the affected model is updated, and the inequality assessment results are recalculated based on dynamic accessibility; Based on the dynamic interactive relationship between supply-side capacity, demand-side weight and connection-side accessibility, a quantitative indicator of inequality in access to spatial resources is generated. Combined with the quantitative indicators of inequality, a new resource location scheme is generated through a particle swarm algorithm, which satisfies the minimum spacing constraint, service coverage constraint and functional diversity constraint.

2. The method for dynamic evaluation of inequality in urban resource access based on reachable time constraints according to claim 1 is characterized by: The ecological quality indicator is the Remote Sensing Ecological Index (RSEI), which is generated by the following steps: Extract greenness, humidity, dryness and heat indexes from remote sensing images, including: The greenness index is calculated by the reflectance of the near-infrared band and the red band; The humidity index is calculated by the humidity component of the tasseled cap transformation; Normalizing the greenness, humidity, dryness and heat indices; A comprehensive score is generated based on principal component analysis to quantify resource supply capacity.

3. The method for dynamic evaluation of inequality in urban resource access based on reachable time constraints according to claim 1 is characterized by: The dynamic interaction relationship is calculated through the following method to calculate the dynamic reachability: Defining the supply and demand matching POD ij and TOD ij The product of is the comprehensive barrier factor: PTOD ij =UNDER ij ×TOD ij According to the dynamically adjusted time threshold T th , calculate the attenuation coefficient Lod through the Gaussian function ij ,satisfy: Based on the attenuation coefficient, calculate the comprehensive service value of the demand point 4. The method for dynamic evaluation of inequality in urban resource access based on reachable time constraints according to claim 1 is characterized by: The data of the affected models are updated in the following situations: When a new supply point is added, only the supply capacity data of the supply-side model is updated, and the dynamic accessibility with the affected demand points is recalculated; When the population distribution on the demand side changes, only the demand weight data of the demand side model is updated, and the reachable time of the connection end is re-matched; When the traffic network is optimized, only the path time data of the connection end is updated through the online map API, and the decay function is recalculated.

5. The method for dynamic evaluation of inequality in urban resource access based on reachable time constraints according to claim 1 is characterized by: The objective function of the particle swarm algorithm is: Where: F is the total cost; represents the distance cost term, which is used to minimize the sum of the reachable distances; αS is the service coverage reward term, which is used to maximize the service coverage of the newly built park; P and λ represent the penalty term and the corresponding penalty weight, respectively, which are used to deal with the violation of the constraint conditions; D ij is the Euclidean distance from demand point i to newly built resource j; S is the number of demand points covered by the newly built resources; P unreach 、P no_serve 、P new 、P exist These are the penalty items for demand point unreachable, newly built resources not serving, insufficient spacing between newly built resources, and insufficient spacing between newly built resources and existing facilities; Constraints include: Newly created resources meet the minimum spacing constraint; The service coverage rate meets the preset threshold; Maintain a minimum safe distance from existing facilities.

6. The method for dynamic evaluation of inequality in urban resource access based on reachable time constraints according to claim 1 is characterized by: The supply-side modeling further includes: combining the functional mix of surrounding points of interest (POIs) to correct the comprehensive score of resource supply capabilities.

7. The method for dynamic evaluation of inequality in urban resource access based on reachable time constraints according to claim 1 is characterized by: The calculation of the Euclidean distance Dij includes: performing path planning on multiple entry points of the same resource, and selecting the entry with the minimum time cost as the valid path.

8. The method for dynamic evaluation of inequality in urban resource access based on reachable time constraints according to claim 1 is characterized by: The calculation of the shortest path time includes: performing path planning on multiple entry points of the same resource, and selecting the entry with the minimum time cost as the valid path.

9. A dynamic evaluation system for inequality in urban resource access based on reachable time constraints, characterized by: include: Supply-side modeling module: used to quantify resource supply capacity through the Remote Sensing Ecological Index (RSEI); Demand-side modeling module: used to generate demand distribution weights based on cellular grid subdivision and population distribution data; Connection end modeling module: Calls the online map API to obtain the real-time shortest path time between supply and demand sides under multiple travel modes, and dynamically calculates the comprehensive supply service volume enjoyed by each demand side based on the dynamic interactive relationship between the supply and demand matching degree (POD) and the reachable time (TOD); Dynamic incremental update module: When any parameter on the supply side, demand side, or connection side changes, it triggers local data update and recalculates the inequality assessment results; Inequality Assessment Module: Generates quantitative indicators of inequality in access to spatial resources based on the dynamic interaction between supply-side capacity, demand-side weight, and connection-side accessibility; Multi-constraint optimization module: Generates new resource location plans that meet minimum spacing constraints, service coverage constraints, and functional diversity constraints through particle swarm optimization.

10. A dynamic evaluation system for implementing the method according to any one of claims 1 to 8, characterized in that: include: Data input interface, used to access remote sensing images, population distribution and traffic network data; a processor configured to perform supply-side modeling, demand-side modeling, connection-side modeling, dynamic incremental updating, inequality assessment, and multi-constraint optimization algorithms; Output interface for generating resource optimization site selection plans and inequality visualization reports.

Images