Historical and cultural block spatial evolution simulation method and system based on GIS space-time analysis

Abstract

The invention provides a historical and cultural block spatial evolution simulation method and system based on GIS space-time analysis, and the method comprises the steps: firstly obtaining a target space-time data set of a target block, which comprises geographic space distribution, building form change, humanistic activity tracks and the like, and then carrying out the preprocessing of a space-time topological structure; generating a target spatio-temporal topological data set containing a spatio-temporal association block space unit sequence and a dynamic attribute tag, and then extracting a static structure, a dynamic evolution mode and a space interaction dependency feature of each space unit based on a preset spatio-temporal evolution feature extraction model; and calling a space-time evolution prediction model to obtain evolution probability distribution and evolution association strength, and finally constructing a space-time evolution path map according to the evolution probability distribution and the evolution association strength, thereby generating a spatial form simulation result and planning intervention decision parameters of a target block, and realizing effective simulation and planning decision support for spatial evolution of historical and cultural blocks.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular, to a method and system for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis. Background Art

[0002] Historical and cultural blocks carry the historical memory and cultural heritage of the city. The research on their spatial evolution is crucial for the protection and utilization of historical and cultural blocks. However, existing technologies mostly focus on the static description of the current situation of the blocks or only conduct simple analysis on a single factor, making it difficult to capture the dynamic process and internal laws of the spatial evolution of the blocks. For the complex interaction relationships between the spatial units of the blocks, there are also lack of effective analysis means, and it is impossible to accurately depict the mutual influence of the spatial units during the evolution process. Moreover, most of the existing spatio-temporal evolution prediction models are based on simple assumptions or limited data, making it difficult to handle the complexity and uncertainty in the spatial evolution of historical and cultural blocks, resulting in low accuracy of the prediction results and unable to provide a reliable decision-making basis for the planning intervention of the blocks.

[0003] Therefore, existing technologies are difficult to comprehensively and accurately simulate the spatial evolution process of historical and cultural blocks and cannot meet the actual needs of the protection and development of the blocks. Summary of the Invention

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present invention, embodiments of the present invention provide a method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis, and the method includes: Obtain a target spatio-temporal data set of a target block, where the target spatio-temporal data set includes geographical spatial distribution data, building form change data, and human activity trajectory data in historical periods; Perform spatio-temporal topological structure preprocessing on the target spatio-temporal data set to generate a target spatio-temporal topological data set, where the target spatio-temporal topological data set includes a sequence of block spatial units with spatio-temporal association and dynamic attribute tags of each spatial unit in the sequence of block spatial units; Based on a preset spatio-temporal evolution feature extraction model, perform feature extraction on the target spatio-temporal topological data set to obtain static structure features, dynamic evolution mode features, and spatial interaction dependence features of each spatial unit; Call a spatio-temporal evolution prediction model to perform spatio-temporal evolution prediction on the static structure features, dynamic evolution mode features, and spatial interaction dependence features to generate an evolution probability distribution of the spatial unit and an evolution correlation intensity between adjacent spatial units; Construct a spatio-temporal evolution path map based on the evolution probability distribution and the evolution correlation intensity, and generate a spatial form simulation result and planning intervention decision parameters of the target block based on the spatio-temporal evolution path map.

[0005] On the other hand, an embodiment of the present invention further provides a historical and cultural block space evolution simulation system based on GIS spatio-temporal analysis, including a processor and a machine-readable storage medium. The machine-readable storage medium is connected to the processor. The machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.

[0006] Based on the above aspects, the embodiment of the present invention integrates the geographical space distribution data, building form change data and human activity trajectory data in historical periods to construct a target spatio-temporal data set, preprocesses the spatio-temporal topological structure of the target spatio-temporal data set, and generates a target spatio-temporal topological data set including a sequence of block space units with spatio-temporal associations and dynamic attribute tags, realizing a deep analysis of the complex relationships in the block space, converting the originally discrete data into structured information with internal logical associations, and enabling the evolution law of the block space to be clearly presented in the spatio-temporal dimension. Feature extraction is carried out based on a preset spatio-temporal evolution feature extraction model, which can accurately extract the static structure features, dynamic evolution mode features and spatial interaction dependence features of each space unit. It not only reflects the inherent attributes of the space unit itself, but also reveals its change trend during the passage of time and its interaction relationship with surrounding space units. Then, a spatio-temporal evolution prediction model is called to predict the spatio-temporal evolution of the extracted features, generating the evolution probability distribution of the space unit and the evolution correlation strength between adjacent space units, further quantifying the evolution possibility and mutual influence degree of the block space from the perspectives of probability and correlation, making the evolution process predictable and quantifiable. Finally, a spatio-temporal evolution path map is constructed according to the evolution probability distribution and evolution correlation strength, and based on this, a spatial form simulation result and planning intervention decision parameters of the target block are generated. This spatio-temporal evolution path map is a comprehensive and intuitive display of the block space evolution process, the spatial form simulation result provides a visual reference for the future development of the block, and the planning intervention decision parameters provide an accurate basis for the scientific planning and reasonable protection of the block. Thus, a comprehensive simulation and accurate prediction of the historical and cultural block space evolution process are realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 is a schematic execution flow diagram of a historical and cultural block space evolution simulation method based on GIS spatio-temporal analysis provided by an embodiment of the present invention.

[0008] Figure 2 is a schematic diagram of exemplary hardware and software components of a historical and cultural block space evolution simulation system based on GIS spatio-temporal analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0009] The embodiments of the present invention will be specifically described below in conjunction with the accompanying drawings of the specification. Figure 1 It is a schematic flowchart of a method for simulating the spatial evolution of a historical and cultural block based on GIS spatio-temporal analysis provided by an embodiment of the present invention. The steps involved will be introduced in detail below.

[0010] Step S110: Obtain a target spatio-temporal data set of the target block, where the target spatio-temporal data set includes geographical spatial distribution data, building form change data, and human activity trajectory data in historical periods.

[0011] In this embodiment, the geographical spatial distribution data, building form change data, and human activity trajectory data in historical periods are within the scope of protection of the historical and cultural block, and the scope of protection of the historical and cultural block, the boundary of the core protection area, and the boundary of the construction control zone need to be combined during the data collection process. By focusing on the scope of protection of the historical and cultural block, interference from irrelevant geographical information can be avoided, making the description of the relative position relationship and layout of buildings, streets, etc. more realistic. The building form change data within this scope can clearly sort out the evolution context of the unique building style of the block, preventing analysis deviation caused by mixing in irrelevant building information. The human activity trajectory data is bounded by the scope of protection, which can truly restore the historical and cultural scene of the block and avoid interference from activity information in other areas.

[0012] The boundary of the core protection area delimits the area with the most historical and cultural value in the block. The data within this area can reflect the core characteristics and original features of the block. In-depth research on it helps to accurately protect historical and cultural heritage. The boundary of the construction control zone is a buffer area for protection. Its data can reflect the impact of the surrounding environment on the core area and provide a basis for formulating reasonable protection strategies, ensuring that the block can retain its historical charm while adapting to modern needs during the development process.

[0013] Among them, the geographical spatial distribution data, building form change data, and human activity trajectory data in historical periods can all be directly retrieved from a pre-constructed resource database. Specifically, the construction process of this resource database is not specifically limited. Here, an exemplary implementation manner will be described in this embodiment, but it should not be understood as a limitation on obtaining the target spatio-temporal data set of the target block.

[0014] Specifically, in the process of constructing geospatial distribution data, historical documents, old maps, photos, and on-site investigations can be used to meticulously restore the geospatial information of the target block in different historical periods. For example, by retrieving historical archival materials, the relative positional relationships and general layout descriptions of various buildings, streets, and public areas in the block at different historical times can be obtained. For instance, regarding Building X in the block, its approximate location and shape in 1850 are known from historical document records. At the same time, referring to the urban planning drawings of that time, its topological adjacency relationships with surrounding buildings Y and Z are determined. For example, Building X and Building Y are adjacent along a certain side. For the vertex coordinate information of the building, the current situation of the block can be surveyed using geospatial mapping tools. Then, based on the relative positional relationships and scale information in historical materials, through spatial geometric calculations and coordinate conversions, the approximate vertex coordinates of the building in different historical periods can be deduced. For example, in modern surveying, with a certain fixed reference point as the benchmark, the vertex coordinates of the current situation of Building X are determined. Then, according to the description of the position changes of Building X in different periods and the scale relationship in historical materials, it is estimated that its vertex coordinates in 1850 are approximately (10, 20), (30, 20), (30, 40), (10, 40). As time progresses, for key time nodes such as 1900 and 1950, the above data collection and deduction processes are repeated to obtain complete geospatial distribution data for different periods.

[0015] In the process of constructing the building form change data, the historical evolution of the building form can be sorted out by retrieving historical building drawings, engineering archives, local chronicles and other materials, as well as on-site investigation of the existing structure and decorative details of the building, in combination with the methods of architectural chronology and style analysis. Specifically, in terms of building density, by recording the number of buildings in the block in different historical periods in historical materials and estimating the block area, the building density of a specific area in different periods can be calculated. For example, the number of buildings in a specific area of the block in 1850 is counted from historical documents, and the area of this area is estimated according to the map materials at that time, resulting in a building density of 40 buildings per square kilometer; by 1900, after obtaining the number of buildings and re-estimating the area through a similar method, the building density has increased to 55 buildings per square kilometer, and a detailed building density change sequence is formed by recording year by year. In terms of building height, by using historical photos, written records and the analysis of the existing building structure, the height of each building at different times can be inferred. For example, through the proportional relationship between the building and surrounding objects in historical photos, combined with the description of the building height in written records, it is determined that the tallest building in the block in 1850 was 15 meters, and some newly built buildings reached 25 meters in 1900. The height of each building at different times is accurately recorded to form a building height change sequence. In terms of functional zoning, study the land use records, commercial registration materials and residents' life records in historical periods, and record in detail the time and specific area range of the functional transformation. For example, it is found from historical commercial archives that a certain area was mainly registered for residential use in 1850, and there are records of a large number of commercial activities in some areas by 1900, thus determining that the function of this area has gradually changed from residential to a commercial functional area, generating a functional zoning conversion sequence.

[0016] In the process of constructing human activity trajectory data, historical documents, diaries, travelogues, oral history and other materials can be mined and analyzed. For example, taking important events and activities in the block as clues, through a detailed interpretation of historical documents, the action routes and activity scopes of people in the block at different times can be sorted out. For example, by studying materials such as local chronicles and celebrity diaries, it is learned that in a specific historical period, the locations and paths where people's gatherings, commercial activities, etc. in the block concentrated. At the same time, using historical maps and spatial analysis methods, the above activity information is located and marked in the geographical space. For the duration of people's stay at different locations, by analyzing information such as the duration of activities and the order of events in historical materials, combined with reasonable speculation and estimation, the approximate stay time is obtained. Taking a week as a cycle, the human activity information collected at different times is sorted out and summarized to accumulate a large amount of data. Then, the above trajectory data is processed by an algorithm to be spatially rasterized, and the block is divided into many small raster cells. For example, in a specific time period, a large number of people are monitored to be frequently active in a raster cell in the center of the block, showing a high activity intensity area on the human activity heat distribution map with a darker color; while only a small number of people pass by a raster cell at the edge, showing a low activity intensity area with a lighter color. At the same time, the time series curve of the activity intensity of each spatial unit is extracted. For example, the activity intensity of a certain spatial unit is low from 9 am to 11 am, reaches the peak from 12 pm to 1 pm, and then gradually decreases, forming the time series curve of the activity intensity of this unit. And the distribution ratio of different activity types in this spatial unit is counted, such as leisure activities accounting for 25%, commercial activities accounting for 45%, and residence-related activities accounting for 30%.

[0017] Step S120: Perform spatio-temporal topological structure preprocessing on the target spatio-temporal data set to generate a target spatio-temporal topological data set, where the target spatio-temporal topological data set includes a sequence of block spatial units with spatio-temporal association and dynamic attribute labels of each spatial unit in the sequence of block spatial units.

[0018] In this embodiment, step S120 may include: Step S121: Perform coordinate calibration processing on the geospatial distribution data to obtain spatial unit boundary data in a unified coordinate system, where the spatial unit boundary data includes the geometric shape vertex coordinates of each spatial unit and the topological adjacency relationship.

[0019] In this embodiment, since the geospatial distribution data obtained in different periods may be based on different coordinate systems, it is necessary to calibrate their coordinates. For example, the local independent coordinate system was used in early surveys and mapping, and the WGS84 coordinate system was used later. Therefore, coordinate conversion software can be selected to convert all geospatial distribution data to the WGS84 coordinate system according to the corresponding conversion algorithm. For example, for the vertex coordinates (10, 20), (30, 20), (30, 40), (10, 40) of Building X in the local independent coordinate system before, after conversion, they become (100.123, 200.456), (100.345, 200.456), (100.345, 200.678), (100.123, 200.678) in the WGS84 coordinate system. At the same time, reconfirm its topological adjacency relationship with the surrounding Buildings Y and Z to ensure the accuracy of the adjacency relationship in the new coordinate system.

[0020] Step S122: Perform time series alignment processing on the building form change data to obtain a building density change sequence, a building height change sequence, and a functional zoning conversion sequence that match the spatial unit boundary data.

[0021] In this embodiment, since the time intervals for obtaining the building form change data in different stages are inconsistent, time series alignment is required. For example, the building density data was obtained in 1850, 1900, and 1950, while the building height data was obtained in 1860, 1910, and 1960. By methods such as linear interpolation, align the building height data to the same time nodes as the building density data. For the time point of 1900, the original building height data was missing. By performing linear interpolation calculations on the data of 1860 and 1910, the approximate building height data for 1900 was obtained, so that the building density change sequence and the building height change sequence match the spatial unit boundary data in time, and at the same time ensure that the functional zoning conversion sequence also accurately corresponds on the same time scale.

[0022] Step S123: Perform spatial rasterization processing on the human activity trajectory data to generate a human activity heat distribution map corresponding to the spatial unit boundary data, and extract the activity intensity time series curve and activity type distribution ratio of each spatial unit in the human activity heat distribution map.

[0023] In this embodiment, spatial analysis software can be used to perform spatial rasterization processing on the human activity trajectory data. For example, a block can be divided into square grid cells with a side length of 10 meters, and then the number of people passing through each grid cell within a specific time period can be counted. For example, within one hour, 50 people passed through grid A at the center of the block, while only 5 people passed through grid B at the edge. Based on the above data, a human activity heat map is generated. Grid A is shown as a high-activity-intensity area with a red color; grid B is shown as a low-activity-intensity area with a light blue color. From the heat map, the activity intensity time series curve of each spatial unit is extracted. For example, the activity intensity of grid C is 10 person-times per hour from 7 am to 9 am, increases to 30 person-times per hour from 10 am to 12 pm, and then gradually decreases. At the same time, the distribution ratio of different activity types within each spatial unit is counted. For example, within grid D, leisure activities account for 35%, commercial activities account for 40%, and residence-related activities account for 25%.

[0024] Step S124: Perform spatio-temporal topological association coding on the spatial unit boundary data, building density change sequence, building height change sequence, functional partition conversion sequence, activity intensity time series curve, and activity type distribution ratio to generate the target spatio-temporal topological data set, where the dynamic attribute labels of each spatial unit include the building density change rate, functional conversion frequency, and activity intensity fluctuation amplitude.

[0025] In this embodiment, taking the spatial unit E as an example, its boundary data in the unified coordinate system, the building density change sequence from 1850 to 1950 (such as the building density in 1850 was 40 buildings per square kilometer, 55 buildings per square kilometer in 1900, and 70 buildings per square kilometer in 1950), the building height change sequence (such as 15 meters in 1850, 25 meters in 1900, and 30 meters in 1950), the functional zoning conversion sequence (such as residential function in 1850, partial commercial function in 1900, and mainly commercial function in 1950), the activity intensity time series curve (such as the activity intensity from 8 am to 10 am was 15 person-times per hour, and 40 person-times per hour from 12 pm to 2 pm), and the activity type distribution ratio (such as 20% for leisure activities, 60% for commercial activities, and 20% for residential-related activities) are integrated and encoded. Calculate the building density change rate of this spatial unit. From 1850 to 1900, in 50 years, the building density increased from 40 buildings per square kilometer to 55 buildings per square kilometer, and the change rate was (55 - 40) ÷ 50 = 0.3 buildings per square kilometer per year; count the frequency of function conversion. From 1850 to 1950, the function changed from residential to commercial, and the conversion frequency was 1 time; calculate the activity intensity fluctuation range. The activity intensity from 8 am to 10 am was 15 person-times per hour, and 40 person-times per hour from 12 pm to 2 pm, and the fluctuation range was 40 - 15 = 25 person-times per hour. Thus, the above dynamic attribute tags are encoded together with other data to form a target spatio-temporal topological data set.

[0026] Step S130: Based on a preset spatio-temporal evolution feature extraction model, extract features from the target spatio-temporal topological data set to obtain the static structure features, dynamic evolution mode features, and spatial interaction dependence features of each spatial unit.

[0027] In this embodiment, step S130 may include: Step S131: Extract the spatial unit boundary data in the target spatio-temporal topological data set, and call a spatial morphology encoder to generate an initial structure feature vector for each spatial unit. The initial structure feature vector includes shape complexity, area change trend, and the number of adjacent units.

[0028] For example, taking the spatial unit F as an example, assume that the boundary data is shown as an irregular polygon with vertex coordinates (100, 100), (120, 110), (130, 130), (110, 140), etc. Call the spatial form encoder to measure the shape complexity by calculating parameters such as the number of sides and interior angles of the polygon. This polygon has 4 sides, and through the set algorithm, the shape complexity is calculated to be 1.2 (the specific algorithm is based on the settings of the spatial form encoder). Also, calculate the area change trend by comparing the areas of this spatial unit at different times. Assume that in the past 50 years, the area has increased from 1000 square meters to 1200 square meters, with an average annual increase of (1200 - 1000) ÷ 50 = 4 square meters, and the area change trend is obtained as 4 square meters / year. At the same time, count the number of adjacent units. The spatial unit F is adjacent to 3 surrounding spatial units. Thus, the shape complexity 1.2, the area change trend of 4 square meters / year, and the number of adjacent units 3 can be spliced to generate the initial structural feature vector.

[0029] Step S132: Perform temporal convolution processing on the building density change sequence, building height change sequence, and functional zoning conversion sequence in the target spatio-temporal topology data set to generate a first dynamic evolution feature vector, where the first dynamic evolution feature vector includes the building density fluctuation period, height growth gradient, and functional conversion trigger threshold.

[0030] For example, the building density change sequence of a certain spatial unit is (40, 45, 50, 55, 60), with a time span of 5 years. Through the temporal convolution algorithm, the building density fluctuation period is calculated to be 3 years. Similar processing is performed on the building height change sequence. For example, the building height sequence is (15, 18, 20, 22, 25), with a time span of 5 years, and the height growth gradient is calculated to be (25 - 15) ÷ 5 = 2 meters / year. For the functional zoning conversion sequence, such as the conversion from residential function to commercial function, analyze the conditions and related factors for the conversion, and determine that the functional conversion trigger threshold is when the proportion of commercial activities reaches 40%, the function begins to convert. Splice the building density fluctuation period of 3 years, the height growth gradient of 2 meters / year, and the functional conversion trigger threshold of 40% to generate the first dynamic evolution feature vector.

[0031] Step S133: Perform spatio-temporal attention processing on the activity intensity time series curve and activity type distribution ratio in the target spatio-temporal topology data set to generate a second dynamic evolution feature vector, where the second dynamic evolution feature vector includes the activity intensity peak phase, type mixing degree, and spatial diffusion direction.

[0032] Taking a certain spatial unit as an example, assuming that the time series curve of its activity intensity shows a peak from 2 pm to 4 pm, the peak phase of the activity intensity is determined to be from 2 pm to 4 pm. The distribution ratio of activity types is statistically analyzed. For example, leisure activities account for 30%, commercial activities account for 50%, and residence-related activities account for 20%. The type mixing degree (the specific algorithm is based on the processing rules) is calculated by setting an algorithm to be 0.6. The activity trajectories of the personnel are analyzed, and the spatial diffusion direction is determined to be from the block center to the periphery. Thus, the peak phase of the activity intensity from 2 pm to 4 pm, the type mixing degree of 0.6, and the spatial diffusion direction from the block center to the periphery can be spliced together to generate the second dynamic evolution feature vector.

[0033] Step S134: Perform cross-scale feature splicing on the initial structure feature vector, the first dynamic evolution feature vector, and the second dynamic evolution feature vector to obtain a static structure feature.

[0034] In this embodiment, the previously generated initial structure feature vector (including the shape complexity of 1.2, the area change trend of 4 square meters per year, and the number of adjacent units of 3), the first dynamic evolution feature vector (including the building density fluctuation period of 3 years, the height growth gradient of 2 meters per year, and the function conversion trigger threshold of 40%), and the second dynamic evolution feature vector (including the peak phase of the activity intensity from 2 pm to 4 pm, the type mixing degree of 0.6, and the spatial diffusion direction from the block center to the periphery) can be subjected to cross-scale feature splicing. According to the set splicing rules, the features of the above different dimensions are combined together to form a comprehensive static structure feature vector, and this static structure feature vector contains multi-faceted structure information of the spatial unit.

[0035] Step S135: Perform graph convolution processing on the topological adjacency relationship in the target spatio-temporal topology data set to generate spatial interaction dependence features, and the spatial interaction dependence features include the density conduction coefficient, the function compatibility index, and the activity diffusion resistance value between adjacent units.

[0036] Step S1351: Construct an undirected graph structure with the spatial units in the spatial unit boundary data as nodes and the topological adjacency relationship as edges.

[0037] In this embodiment, an undirected graph structure can be constructed based on the spatial unit boundary data in the previously processed target spatio-temporal topology data set. Taking multiple spatial units in a block as an example, assume spatial units A, B, C, D, etc. It is known that the topological adjacency relationships between them, such as spatial unit A is adjacent to spatial unit B, spatial unit B is adjacent to spatial unit C, spatial unit C is adjacent to spatial unit D, etc. Thus, the above spatial units can be used as nodes of the undirected graph, and their adjacency relationships can be used as edges to construct an undirected graph structure. This undirected graph structure can intuitively reflect the adjacency relationships between each spatial unit and provide an infrastructure for subsequent feature extraction and analysis. For example, in this undirected graph, spatial unit A, as a node, is connected to the adjacent spatial unit B through an edge, and so on, forming a complete graph structure to describe the topological relationships between the spatial units in the block.

[0038] Step S1352: Initialize the initial feature vectors of each node in the undirected graph structure. The initial feature vectors are generated by splicing the shape complexity, area change trend, and number of adjacent units in the static structure features with the building density fluctuation period and function conversion trigger threshold in the dynamic evolution pattern features.

[0039] In this embodiment, for each node in the constructed undirected graph structure, that is, each spatial unit, the initial feature vectors are initialized. Continuing with the previous example of spatial unit E, the shape complexity in its static structure features is 1.5 (obtained through previous calculations of its boundary data, and this spatial unit is a slightly complex polygon shape), the area change trend is an increase of 3 square meters per year (obtained through analysis of area data at different times), and the number of adjacent units is 4 (it is adjacent to 4 surrounding spatial units). The building density fluctuation period in its dynamic evolution pattern features is 4 years (obtained from the analysis of the building density change sequence), and the function conversion trigger threshold is 50% (the function will change when the proportion of commercial activities reaches 50%). Then, the above features are spliced to form the initial feature vector of spatial unit E. In the same way, corresponding initial feature vectors are generated for each node in the undirected graph structure.

[0040] Step S1353: Perform multi-layer graph convolutional network processing on the undirected graph structure, and layer by layer aggregate the feature vectors of the adjacent nodes of each node to generate intermediate spatial interaction features.

[0041] In this embodiment, the constructed and initialized undirected graph structure with node feature vectors is processed using a multi-layer graph convolutional network. In the first layer of the graph convolutional network, taking the spatial unit F as an example, it has adjacent nodes G, H, and I. The graph convolutional network can aggregate the initial feature vector of the spatial unit F itself and the initial feature vectors of its adjacent nodes G, H, and I according to the set rules. Suppose the initial feature vector of the spatial unit G contains information such as shape complexity 1.3, area change trend 2 square meters per year, number of adjacent units 3, building density fluctuation period 3.5 years, function conversion trigger threshold 45%, etc.; the initial feature vector of the spatial unit H contains corresponding different values; the spatial unit I also has its unique initial feature vector. Through the calculation rules of the graph convolutional network (for example, according to rules such as adjacency relationship and weight assignment of feature vectors), the feature vectors of the above adjacent nodes are aggregated with the feature vector of the spatial unit F to generate the intermediate spatial interaction feature of the spatial unit F after being processed by the first layer of the graph convolutional network. This intermediate spatial interaction feature integrates partial information of the spatial unit F and its adjacent nodes, and initially reflects the interaction relationship between spatial units.

[0042] Then enter the second layer of the graph convolutional network. In the same way, the intermediate spatial interaction feature after the first layer of processing and the features of the adjacent nodes after the first layer of processing can be further aggregated, continuously enriching and updating the intermediate spatial interaction feature, and so on. After being processed by multiple layers of the graph convolutional network, a more comprehensive intermediate spatial interaction feature is obtained.

[0043] Step S1354: After each layer of the graph convolutional network is processed, input the intermediate spatial interaction feature into a gated recurrent unit, generate a neighborhood feature fusion weight based on the function conversion frequency in the dynamic attribute label, perform weighted superposition on the intermediate spatial interaction feature and the current node feature vector, update the node feature vector, and obtain the node feature vector output by the final layer.

[0044] In this embodiment, after each layer of the graph convolutional network finishes processing, the generated intermediate spatial interaction features are input into the gated recurrent unit. Taking the spatial unit J as an example, after a certain layer of the graph convolutional network processes, intermediate spatial interaction features are obtained, and at the same time, it is known that the function conversion frequency in its dynamic attribute label is 2 times (the function has been converted 2 times within a certain time range). The gated recurrent unit can generate neighborhood feature fusion weights according to information such as the function conversion frequency. Assuming that through the calculation rules of the gated recurrent unit, the weight distribution for the feature fusion of the current spatial unit J and its adjacent node features is obtained. For example, for the feature of the adjacent node K, the weight is 0.4; for the feature of the adjacent node L, the weight is 0.3, etc. Then, the intermediate spatial interaction features and the feature vector of the current node (spatial unit J) are weighted and superimposed according to the above weights, that is, the parts of the intermediate spatial interaction features from different adjacent nodes are combined with the feature vector of the spatial unit J itself according to the weights to update the node feature vector of the spatial unit J. After processing through multiple layers of the graph convolutional network and performing such weighted superposition operations for each layer, the node feature vector output by the final layer is finally obtained, which fully integrates the information of the spatial unit itself and its adjacent units after multiple layers of processing, and more comprehensively reflects the characteristics of the spatial unit in spatial interaction.

[0045] Step S1355: Perform spatial scale alignment processing on the node feature vector output by the final layer, encode the spatial positions of the feature vectors of each node with the vertex coordinates of the geometric shape of the spatial unit boundary data, and generate spatial interaction dependence features including density conduction gradients, function compatibility matrices, and activity diffusion direction vectors.

[0046] In this embodiment, since the eigenvectors of different spatial units may vary in scale, unified processing is required. For example, for the final node eigenvector of spatial unit M, through a set algorithm and rules, the numerical values of its respective dimensions are adjusted so that it is comparable to the eigenvectors of other spatial units in terms of spatial scale. Then, the eigenvector of each node is spatially position-encoded with the vertex coordinates of the geometric shape of the spatial unit boundary data. Taking spatial unit N as an example, the vertex coordinates of the geometric shape of its boundary data are (100, 150), (120, 160), etc. The node eigenvector is combined with the above vertex coordinates through a set encoding method (for example, by performing correlation calculations on certain dimensions of the eigenvector and the numerical values of the vertex coordinates). Through such processing, spatial interaction dependence features including density conduction gradients, functional compatibility matrices, and activity diffusion direction vectors are generated. For the density conduction gradient, by analyzing factors such as the building density changes between the spatial unit and its adjacent units and the distances between them, the numerical value of the density conduction gradient from one spatial unit to its adjacent unit is calculated; the functional compatibility matrix is a matrix constructed based on the functional partition conversion between different spatial units and their adjacent relationships to represent the degree of functional compatibility between each spatial unit; the activity diffusion direction vector is to comprehensively determine information such as the diffusion direction and intensity of human activities between spatial units based on the human activity trajectory data and the topological adjacency relationships of the spatial units, and finally form spatial interaction dependence features that comprehensively reflect the interaction dependence relationships between spatial units.

[0047] Step S136: Dynamically weight and correct the static structure features based on the dynamic attribute tags in the target spatio-temporal topology data set to obtain dynamic evolution pattern features.

[0048] In this embodiment, taking the spatial unit P as an example, its static structural features have been generated through the previous steps, including information such as shape complexity and area change trend. At the same time, the dynamic attribute tags of this spatial unit are recorded in the target spatio-temporal topological data set. For example, the building density change rate increases by 2 buildings per square kilometer per year, the function conversion frequency is 3 times (within a specific time period), and the activity intensity fluctuation range is 20 person-times per hour. Based on the above dynamic attribute tags, the static structural features are dynamically weighted and corrected. For example, for the feature of shape complexity, if the function conversion frequency is high, it indicates that the function of the spatial unit changes relatively actively, which may affect its future shape development. Therefore, a relatively high weight, assumed to be 0.6, is assigned to the feature of shape complexity. For the area change trend, the weight is adjusted according to factors such as the building density change rate, and the assumed weight is 0.4. In this way, the weighted calculation is performed on each part of the static structural features, and the weighted results are combined to obtain the dynamic evolution pattern features. The dynamic evolution pattern features take into account both the static structural basis of the spatial unit and its dynamic change attributes, and more accurately reflect the evolution trend of the spatial unit in the time dimension.

[0049] Step S140: Invoke the spatio-temporal evolution prediction model to perform spatio-temporal evolution prediction on the static structural features, dynamic evolution pattern features, and spatial interaction dependence features, and generate the evolution probability distribution of the spatial unit and the evolution correlation strength between adjacent spatial units.

[0050] Step S141: Input the static structural features into the first branch network of the spatio-temporal weight assignment model to generate the internal evolution weight coefficient of the spatial unit, and the internal evolution weight coefficient reflects the influence degree of the historical form on the current evolution direction.

[0051] In this embodiment, the previously extracted static structural features are input into the first branch network of the spatio-temporal weight assignment model. The first branch network will analyze the historical form features of this spatial unit based on the above information. If the shape of this spatial unit has been relatively stable and the area change has also been relatively regular in the past for a long time, then the influence of the historical form on the current evolution direction may be relatively large. This internal evolution weight coefficient indicates that the historical form has a relatively large weight in the future evolution direction of this spatial unit, meaning that its future evolution may be restricted and guided by the past form to a certain extent. Different spatial units will generate corresponding internal evolution weight coefficients through the first branch network according to their respective static structural features, and the above internal evolution weight coefficients reflect the different influence degrees of the historical form of each spatial unit on the current evolution direction.

[0052] Step S142: Input the dynamic evolution pattern features into the second branch network of the spatio-temporal weight allocation model to generate a time decay adjustment coefficient, which reflects the contribution decay rate of data in different periods to the prediction result.

[0053] In this embodiment, the second branch network analyzes the variation of the above-mentioned dynamic evolution pattern features over time. For example, if the fluctuation period of building density is relatively stable, but over time, the influence of early data on the current and future evolution may gradually decrease. Through the algorithm inside the network, combining historical data and time factors, the time decay adjustment coefficient of the spatial unit is calculated. Different spatial units generate their respective time decay adjustment coefficients through the second branch network according to the differences in their dynamic evolution pattern features, which are used to adjust the weights of data in different periods during the prediction of evolution.

[0054] Step S143: Input the spatial interaction dependence features into the third branch network of the spatio-temporal weight allocation model to generate a neighborhood coupling strength coefficient, which reflects the cooperative or competitive relationship of the evolution behaviors of adjacent spatial units.

[0055] In this embodiment, the spatial interaction dependence features can be input into the third branch network of the spatio-temporal weight allocation model. Taking the spatial unit S and its adjacent units T and U as an example, the spatial interaction dependence features include information such as density conduction gradient, functional compatibility matrix, and activity diffusion direction vector. The third branch network analyzes the interaction relationship between the spatial unit S and the adjacent units T and U. If the functional compatibility between the spatial unit S and the adjacent unit T is relatively high, and the diffusion of personnel activities between them is relatively frequent, it indicates that their evolution behaviors may have strong cooperation; on the contrary, if there is functional competition or activity diffusion obstruction, there may be a competitive relationship. Through the analysis and calculation of the above interaction information, the neighborhood coupling strength coefficient of the spatial unit S is generated as 0.6. This neighborhood coupling strength coefficient represents the degree of the cooperative or competitive relationship of the evolution between the spatial unit S and the adjacent units. Different spatial units generate their respective neighborhood coupling strength coefficients through the third branch network according to different situations of their spatial interaction dependence features, so as to reflect their interaction relationship with adjacent spatial units during the evolution process.

[0056] Step S144: Construct a spatio-temporal joint weight matrix according to the internal evolution weight coefficient, time decay adjustment coefficient, and neighborhood coupling strength coefficient, and perform weighted fusion on the static structure features, dynamic evolution pattern features, and spatial interaction dependence features to generate a spatio-temporal fusion feature vector.

[0057] In this embodiment, a spatio-temporal joint weight matrix can be constructed based on the internally evolved weight coefficient, time decay adjustment coefficient, and neighborhood coupling strength coefficient generated previously. Taking the spatial unit V as an example, its internally evolved weight coefficient is 0.7, the time decay adjustment coefficient is 0.8, and the neighborhood coupling strength coefficient is 0.5. Arrange the above coefficients in a matrix form according to the set rules. For example, place the internally evolved weight coefficient in the upper left corner of the matrix, the time decay adjustment coefficient in the middle, and the neighborhood coupling strength coefficient in the lower right corner, etc. (the specific arrangement method is based on the model setting). Then, use this spatio-temporal joint weight matrix to perform weighted fusion on the static structure feature, dynamic evolution pattern feature, and spatial interaction dependence feature. For each element in the static structure feature, such as shape complexity, area change trend, etc., determine its weight in the fusion according to the internally evolved weight coefficient; for the building density fluctuation period, height growth gradient, etc. in the dynamic evolution pattern feature, determine the weight based on the time decay adjustment coefficient; for the density conduction gradient, functional compatibility matrix, etc. in the spatial interaction dependence feature, determine the weight according to the neighborhood coupling strength coefficient. Through weighted calculation, fuse the above different features to generate a spatio-temporal fusion feature vector. This spatio-temporal fusion feature vector synthesizes various features of the spatial unit and their weight information in the spatio-temporal dimension, providing more comprehensive data support for subsequent evolution prediction.

[0058] Step S145: Input the spatio-temporal fusion feature vector into the evolution probability prediction layer to generate the evolution probability distribution of the spatial unit.

[0059] Input the generated spatio-temporal fusion feature vector into the evolution probability prediction layer. Taking the spatial unit W as an example, its spatio-temporal fusion feature vector contains various information after weighted fusion. The evolution probability prediction layer analyzes the relationship between the information contained in this feature vector and the future evolution possibility of the spatial unit. For example, considering the comprehensive information such as the static structure feature, dynamic evolution pattern feature, and spatial interaction dependence feature of the spatial unit W, and combining the historical data and the rules obtained from model training, predict the evolution probability of the spatial unit in different future states. Different spatial units will generate their respective evolution probability distributions through the evolution probability prediction layer. The above evolution probability distribution reflects different possible evolution situations and their corresponding probabilities of each spatial unit in the future.

[0060] Step S146: Input the spatio-temporal fusion feature vector into the correlation strength prediction layer to generate the evolution correlation strength between adjacent spatial units.

[0061] Input the spatio-temporal fusion feature vector into the correlation strength prediction layer. Taking the spatial unit X and its adjacent spatial unit Y as an example, the spatio-temporal fusion feature vector contains various information about themselves and each other. The correlation strength prediction layer will, based on the above information, through specific algorithms and models, analyze the degree of correlation between the spatial units X and Y in the future evolution process. For example, consider factors such as functional compatibility and activity diffusion direction in their spatial interaction dependence characteristics, as well as their respective static structures and dynamic evolution mode characteristics. Different adjacent spatial units will generate the numerical values of their evolution correlation strength through the correlation strength prediction layer, and the above-mentioned evolution correlation strength numerical values reflect the degree of mutual dependence and influence between adjacent spatial units in the evolution process.

[0062] Step S150: Construct a spatio-temporal evolution path map based on the evolution probability distribution and the evolution correlation strength, and generate a spatial form simulation result and planning intervention decision parameters of the target block based on the spatio-temporal evolution path map.

[0063] Step S151: Taking each spatial unit as a node and using the evolution correlation strength as the edge weight, construct an initial spatio-temporal evolution graph structure.

[0064] In this embodiment, based on each spatial unit in the block, each spatial unit is taken as a node. For example, using the evolution correlation strength between adjacent spatial units calculated previously as the edge weight, construct an initial spatio-temporal evolution graph structure to form a preliminary graph structure to represent the evolution correlation relationship between spatial units.

[0065] Step S152: Update the node states of the initial spatio-temporal evolution graph structure according to the evolution probability distribution to generate a spatio-temporal evolution graph with probability weights.

[0066] For example, in the initial spatio-temporal evolution graph structure, the state of each node is updated according to its evolution probability distribution to generate a spatio-temporal evolution graph with probability weights, which not only reflects the evolution correlation strength between spatial units, but also considers the probability of different evolution states of each spatial unit itself.

[0067] Step S153: Perform path search processing on the spatio-temporal evolution graph with probability weights, and extract the edges with probability weights greater than the set weight to form a set of main evolution paths.

[0068] For example, in a possible implementation manner, step S153 may include: Step S1531: Perform bidirectional weight sorting on the edges in the spatio-temporal evolution graph based on the evolution correlation strength to generate an initial path search queue.

[0069] In this embodiment, for the spatio-temporal evolution graph with probability weights, this embodiment analyzes the evolution correlation strength of the connections between each spatial unit. For example, taking the local spatio-temporal evolution graph composed of spatial units F, G, H, I, etc. in a block as an example, the evolution correlation strength between spatial unit F and G is 0.7, the evolution correlation strength between F and H is 0.5, and the evolution correlation strength between F and I is 0.6; the evolution correlation strength between G and H is 0.4, the evolution correlation strength between G and I is 0.3, and the evolution correlation strength between H and I is 0.65.

[0070] Bidirectional weight sorting means arranging the evolution correlation strength of the connections in both descending and ascending orders. For example, when sorting in descending order, first determine the maximum evolution correlation strength value. In the above example, 0.65 between H and I is a relatively large value, followed by 0.7 between F and G, then 0.6 between F and I, and so on. When sorting in ascending order, start with 0.3 between G and I as a smaller value, followed by 0.4 between G and H, and so on.

[0071] Organize the sorted connection information into an initial path search queue. This initial path search queue contains the information of all connections and their corresponding weight values, and is arranged in the order of bidirectional sorting. For example, the part sorted in descending order may be: (H-I, 0.65), (F-G, 0.7), (F-I, 0.6) ……; the part sorted in ascending order may be: (G-I, 0.3), (G-H, 0.4) ……. The initial path search queue generated in this way can provide an ordered data basis for subsequent path searches, facilitating the exploration of possible evolution paths according to different weight orders.

[0072] Step S1532: Extract the current maximum-weight connection from the initial path search queue, perform a depth-first traversal along the nodes connected by this connection, and record the cumulative evolution probability of the traversal path and the sequence of spatial units passed through.

[0073] In this embodiment, first extract the current maximum-weight connection from the initial path search queue. Assume that in the current queue, (F-G, 0.7) is the maximum-weight connection. Starting from spatial unit F as the starting node, follow this connection to reach the connected spatial unit G, and start a depth-first traversal.

[0074] During the traversal process, it is necessary to record the cumulative evolution probability of the traversal path. The calculation of the cumulative evolution probability is based on the evolution probability distribution of each spatial unit and the weight of the connection. For example, the probability of spatial unit F evolving to spatial unit G in its evolution probability distribution is 0.5, and the weight of the connection between F and G is 0.7, then the cumulative evolution probability of this step from F to G is the product of the two, that is, 0.5×0.7 = 0.35.

[0075] After reaching the spatial unit G, continue to check the adjacent connecting edges of node G. Suppose the weight of the connecting edge between G and H is 0.4, and the probability of G evolving towards H is 0.6 in its evolution probability distribution. Then the cumulative evolution probability from G to H becomes 0.084, and at the same time, record the sequence of spatial units passed as F - G - H.

[0076] According to the principle of depth - first traversal, continuously explore along the adjacent connecting edges of the current node, and continuously update the cumulative evolution probability and the sequence of spatial units passed until the path cannot be extended any further. For example, during the exploration process, some nodes may be encountered where all their adjacent connecting edges have been visited, or the weights of some connecting edges are lower than the set threshold. In this case, stop the traversal in that direction.

[0077] Step S1533: When traversing to the end node, compare the cumulative evolution probability with the path threshold. If the cumulative evolution probability exceeds the path threshold, mark the sequence of spatial units as a candidate path.

[0078] Suppose the path threshold set in this embodiment is 0.1. When the depth - first traversal reaches the end node, taking the path from F through G to H obtained previously as an example, its cumulative evolution probability is 0.084, which does not exceed the path threshold of 0.1. So this path is not temporarily marked as a candidate path.

[0079] Continue the traversal. Suppose another path starts from spatial unit A, passes through spatial units B, C, and reaches spatial unit D. During the traversal process, the calculated cumulative evolution probability is 0.15. When reaching the end node D, compare the cumulative evolution probability of 0.15 with the path threshold of 0.1. Since 0.15 is greater than 0.1, mark the sequence of spatial units A - B - C - D as a candidate path.

[0080] Through the above comparison process, it is possible to screen out those sequences of spatial units with relatively high cumulative evolution probabilities and a greater likelihood of becoming the main evolution path.

[0081] Step S1534: Perform spatial topological continuity verification on the candidate path, detect whether the adjacent spatial units in the candidate path satisfy the condition of overlapping geometric shape vertex coordinates in the topological adjacency relationship, and eliminate the broken paths according to the detection results to obtain the verified candidate path.

[0082] For the sequence of spatial units marked as candidate paths, taking A - B - C - D as an example, spatial topological continuity verification is carried out. First, check the topological adjacency relationship between spatial units A and B to see if there is an overlapping part in the vertex coordinates of their geometric shapes. Suppose the boundary vertex coordinates of spatial unit A are (100, 100), (120, 100), (120, 120), (100, 120), and the boundary vertex coordinates of spatial unit B are (120, 100), (140, 100), (140, 120), (120, 120). It can be found that they have some overlapping vertex coordinates, satisfying the topological adjacency relationship.

[0083] Next, check the topological adjacency relationship between spatial units B and C. If there is also an overlapping part between the boundary vertex coordinates of spatial unit C and those of B, the adjacency condition is satisfied. Then check the adjacency relationship between C and D. If the condition is also met, the candidate path A - B - C - D passes the spatial topological continuity verification.

[0084] However, if there is a candidate path, such as E - F - G, where there is no overlapping part in the vertex coordinates of spatial units E and F, not satisfying the topological adjacency relationship, then this candidate path is a broken path and needs to be removed.

[0085] By performing such spatial topological continuity verification on all candidate paths and removing the broken paths that do not meet the conditions, the verified candidate paths are obtained. These verified candidate paths are topologically continuous in space and more in line with the actual logic of block space evolution.

[0086] Step S1535: Sort the verified candidate paths in descending order according to the cumulative evolution probability, and select the top N paths as the main evolution path set.

[0087] Suppose after spatial topological continuity verification, multiple verified candidate paths are obtained. For example, the cumulative evolution probability of path P1 is 0.2, the cumulative evolution probability of path P2 is 0.18, the cumulative evolution probability of path P3 is 0.16, and so on.

[0088] Then, sort the verified candidate paths in descending order according to the cumulative evolution probability, that is, in the order from largest to smallest. The sorted order may be path P1, path P2, path P3, etc.

[0089] According to the actual requirements, select the top N paths as the main evolution path set. Suppose N = 3, then the selected main evolution path set includes path P1, path P2, and path P3. This main evolution path set represents the path combinations with a greater possibility of block space evolution under the current analysis and provides core data for the subsequent generation of the spatio - temporal evolution path atlas.

[0090] Step S154: Perform loop detection on the set of main evolution paths. If a closed loop is detected, decay and correct the evolution probability within the loop according to the time decay adjustment coefficient.

[0091] In the set of main evolution paths, take the path containing spatial units M, N, O, and P as an example to check if there is a closed loop. After analysis, it is found that there is a closed loop situation of M - N - O - P - M in the path.

[0092] It is known that each spatial unit has a corresponding time decay adjustment coefficient. Assume that the time decay adjustment coefficient of spatial unit M is 0.8, N is 0.75, O is 0.85, and P is 0.7.

[0093] For the evolution probability within the loop, take the evolution probability from M to N as an example. Assume the original evolution probability is 0.4. Due to the existence of the loop, according to the time decay adjustment coefficient, the corrected evolution probability is 0.24. Using the same method, correct the evolution probabilities of other connected edges within the loop. By decaying and correcting the evolution probability within the loop according to the time decay adjustment coefficient, it can more accurately reflect the actual situation of the block space evolution and avoid the problem of unreasonable amplification of the evolution probability caused by the loop.

[0094] Step S155: Superimpose and render the set of main evolution paths after decay correction with the spatial unit boundary data to generate the spatio-temporal evolution path atlas, which includes a historical form retention area, a high-probability evolution area, and a neighborhood co-evolution corridor.

[0095] In this embodiment, combine the set of main evolution paths after decay correction with the spatial unit boundary data. Based on the geographical information data of the block, draw the path information in the set of main evolution paths onto the map represented by the spatial unit boundary data.

[0096] For example, for the historical form retention area, in the spatial unit boundary data, the building forms in some areas have not changed significantly over a long period of time, and these areas are marked as historical form retention areas in the spatio-temporal evolution path atlas. Assume that the area where spatial unit Q is located has remained relatively stable in terms of building density, building height, and functional zoning in the past few decades, then this area is classified into the historical form retention area.

[0097] The high-probability evolution area is determined according to the cumulative evolution probability of each path in the set of main evolution paths. The areas of spatial units passed by those paths with higher cumulative evolution probabilities are identified as high-probability evolution areas. For example, the areas of spatial units R, S, T, etc. passed by path P1 become high-probability evolution areas because the cumulative evolution probability of path P1 is relatively high, indicating that there is a greater possibility of spatial form evolution in these areas in the future.

[0098] The neighborhood co-evolution corridor is determined based on the adjacency relationship between spatial units and the intensity of evolution association. In the set of main evolution paths, the areas connected by the edges with a relatively high intensity of evolution association between adjacent spatial units form the neighborhood co-evolution corridor. For example, if the intensity of evolution association between spatial units U and V is relatively high and they often appear simultaneously in the main evolution path, the area between them constitutes the neighborhood co-evolution corridor, indicating that these areas may co-evolve and influence each other during future evolution processes.

[0099] By superimposing and rendering the above information, an intuitive spatio-temporal evolution path map is generated, which can clearly display the morphological change trends of the block during the historical development process and the possible future evolution directions.

[0100] Step S156: Mark the spatial unit sequence of the set of main evolution paths in the spatio-temporal evolution path map, and superimpose and match the path intersection nodes with the peak activity intensity regions in the human activity heat distribution map to generate a set of key evolution intervention node coordinates.

[0101] In this embodiment, on the generated spatio-temporal evolution path map, the spatial unit sequence of the set of main evolution paths is clearly marked. For example, for path P1 in the set of main evolution paths, which passes through spatial units A - B - C - D, the above spatial units and the connection paths between them are clearly marked on the spatio-temporal evolution path map.

[0102] Meanwhile, the path intersection nodes are superimposed and matched with the human activity heat distribution map. Taking path intersection node E as an example, in the human activity heat distribution map, the peak activity intensity region corresponding to the position of this node is searched. Assuming that in the human activity heat distribution map, the activity intensity at the position where node E is located reaches the peak during a certain time period, it indicates that this area is important in terms of human activities.

[0103] By superimposing and matching all path intersection nodes with the peak activity intensity regions in the human activity heat distribution map, the key positions that are both on the main evolution path and have a relatively high human activity intensity are determined. The coordinates of the above positions are collected to generate a set of key evolution intervention node coordinates. For example, after matching, the set of key evolution intervention node coordinates obtained is {(x1, y1), (x2, y2), (x3, y3) ……}. The above coordinate points are very important for urban planners because they represent the positions that need to be focused on and may be intervened in during the spatial evolution process of the block, which helps to formulate more reasonable planning strategies.

[0104] Step S157: Extract the set of spatial units in the historical form retention area in the spatio-temporal evolution path atlas, calculate the variance value of the building density change sequence and the mean value of the function conversion frequency therein, and generate a form stability index.

[0105] Accurately extract the set of spatial units in the historical form retention area from the spatio-temporal evolution path atlas. Suppose this set of spatial units includes spatial units X, Y, Z, etc. For each spatial unit, collect the data of its building density change sequence. For example, the building density data of spatial unit X in the past 50 years is (30, 32, 31, 33, 32), the building density data of spatial unit Y is (40, 41, 40, 42, 41), and the building density data of spatial unit Z is (25, 26, 25, 27, 26).

[0106] Then, calculate the variance value of the above building density change sequence. Taking spatial unit X as an example, first calculate the average value according to (30, 32, 31, 33, 32), and then calculate the variance. Similarly, calculate the variance values of spatial units Y and Z.

[0107] Next, calculate the mean value of the function conversion frequency, and synthesize the variance value of the building density change sequence and the mean value of the function conversion frequency obtained above to generate a form stability index. This form stability index can reflect the stability degree of the spatial units in the historical form retention area.

[0108] Step S158: Extract the set of spatial units in the high-probability evolution area in the spatio-temporal evolution path atlas, and generate the maximum probability evolution direction and the expected form conversion type according to its evolution probability distribution.

[0109] Find out the set of spatial units in the high-probability evolution area from the spatio-temporal evolution path atlas, suppose it includes spatial units M, N, O, etc. For each spatial unit, check its evolution probability distribution. For example, the evolution probability distribution of spatial unit M shows that the probability of evolving into an expansion form is 0.6, the probability of evolving into a contraction form is 0.3, and the probability of maintaining the existing form is 0.1; the probability of spatial unit N changing its function to commercial is 0.7, and the probability of maintaining the existing function is 0.3; the probability of spatial unit O increasing the building height is 0.8, and the probability of decreasing the building height is 0.2.

[0110] According to the above evolution probability distribution, determine the maximum probability evolution direction of each spatial unit. For spatial unit M, the maximum probability evolution direction is the expansion form; for spatial unit N, the maximum probability evolution direction is the function conversion to commercial; for spatial unit O, the maximum probability evolution direction is the increase of the building height.

[0111] Meanwhile, summarize the expected form conversion types. Considering the above-mentioned spatial units, the expected form conversion types include form expansion, function transformation (such as residential to commercial), building height change (increase), etc. The above-mentioned most probable evolution directions and expected form conversion types provide important information for planners to understand the development trends of high-probability evolution areas, which helps to formulate corresponding planning strategies in advance.

[0112] Step S159: Extract the edge weight distribution of the neighborhood co-evolution corridors in the spatio-temporal evolution path map, and generate a corridor width adjustment parameter and a function mixing degree improvement coefficient.

[0113] In this embodiment, the edge weight distribution of the neighborhood co-evolution corridors in the spatio-temporal evolution path map can be analyzed. Assume that the neighborhood co-evolution corridor includes the edges between spatial units A and B, B and C, and C and D, and their edge weights are 0.7, 0.8, and 0.75 respectively.

[0114] Then, generate a corridor width adjustment parameter according to the edge weight distribution. For example, through a certain calculation method (assuming according to the average value or weighted average value of the weights, etc.), the corridor width adjustment parameter is calculated. Assuming it is calculated by the average value, (0.7 + 0.8 + 0.75) ÷ 3 = 0.75. This parameter can be used to guide the adjustment of the width of the neighborhood co-evolution corridor in the planning. A higher weight may mean that a wider corridor is needed to promote the co-development between neighborhoods. Assume that according to past experience and planning requirements, when the corridor width adjustment parameter is between 0.7 and 0.8, it is recommended to set the corridor width to 15 meters (this is just an assumed corresponding relationship, and it actually needs to be determined according to specific planning standards).

[0115] For the function mixing degree improvement coefficient, in this embodiment, the functions of the spatial units connected by the connecting edges are further analyzed. Assume that the main function of spatial unit A is residential, the main function of spatial unit B is commercial, and the edge weight between the two is 0.7; the edge weight of the connection between spatial unit B and spatial unit C with the function of leisure is 0.8. By analyzing the functional differences of the above adjacent spatial units and the edge weights, the function mixing degree improvement coefficient is calculated. For example, in this embodiment, a calculation method is set, considering factors such as the edge weight and the complexity of the functional type differences (assuming that the functional types are divided into three types: residential, commercial, and leisure, and different functional type differences are assigned different values. Here, the difference between residential and commercial is assigned a value of 2, the difference between commercial and leisure is assigned a value of 1.5, the difference between residential and leisure is assigned a value of 1, etc., and the specific assignment is determined according to the actual planning considerations). For the connection between A and B, the contribution value to the improvement of the function mixing degree is 0.7×2 = 1.4; for the connection between B and C, the contribution value to the improvement of the function mixing degree is 0.8×1.5 = 1.2. By comprehensively calculating the above contribution values (assuming simple addition), the function mixing degree improvement coefficient is obtained as 1.4 + 1.2 = 2.6. This coefficient reflects the potential of the neighborhood collaborative evolution corridor in promoting the integration of different functional spatial units. The higher the coefficient, the greater the possibility and demand for function mixing. In the planning, the land use planning and facility configuration, etc. can be adjusted accordingly to improve the function mixing degree and enhance the vitality and diversity of the block.

[0116] Step S1510: Construct a multi-dimensional decision parameter set according to the morphological stability index, the maximum probability evolution direction, the expected morphological conversion type, the corridor width adjustment parameter, and the function mixing degree improvement coefficient.

[0117] In this embodiment, the various parameters calculated previously can be integrated, and the integrated multi-dimensional decision parameter set covers multiple key information on the spatial evolution of the target block.

[0118] Step S1511: Input the multi-dimensional decision parameter set into the planning strategy generation model, and output the spatial form simulation result including the spatial control intensity level, the update time sequence priority, and the function adaptation optimization plan.

[0119] In this embodiment, the constructed multi-dimensional decision parameter set can be input into the planning strategy generation model. For the spatial control intensity level, it can be determined according to parameters such as the morphological stability index and the maximum probability evolution direction. For example, if the morphological stability index is low, it indicates that the historical morphology of this area has changed greatly, and a relatively high spatial control intensity may be required to guide its development in a reasonable direction; for areas within the high-probability evolution zone and with a relatively clear maximum probability evolution direction, the spatial control intensity may need to be set specifically. Assume that according to the model setting, when the morphological stability index is lower than 0.6, the spatial control intensity level is set to high, and strict restrictions are required on aspects such as building height and building density in the planning; when the morphological stability index is between 0.6 and 0.8, the spatial control intensity level is medium, and the restrictions are relatively loose; when the morphological stability index is higher than 0.8, the spatial control intensity level is low. In this embodiment, the morphological stability index is 0.57, so the model outputs that the spatial control intensity level of this area is high.

[0120] The determination of the update time sequence priority comprehensively considers factors such as the maximum probability evolution direction and the expected morphological conversion type. For areas with high-probability evolution and where the expected morphological conversion has a greater impact on the development of the block, a higher update time sequence priority will be given. For example, for those spatial units whose maximum probability evolution direction is a functional transformation and involves important functional areas (such as from residential to commercial center), the model will consider that their update time sequence priority is high and they need to be planned and renovated first. Assume that in this embodiment, the areas within the high-probability evolution zone where the function changes from residential to commercial center are determined by the model to have a high update time sequence priority, while the update time sequence priority of some areas with relatively low evolution probability and less impact on the overall function of the block is relatively low.

[0121] The functional adaptation optimization plan is formulated based on the corridor width adjustment parameter, the functional mixing degree improvement coefficient, and the current functional status and expected evolution direction of each spatial unit. For example, according to the corridor width adjustment parameter of 0.75, combined with the planning standard, the road is widened to 15 meters in the neighborhood collaborative evolution corridor area to promote the circulation of people and activities and strengthen functional mixing. For the case where the functional mixing degree improvement coefficient is 2.6, the model recommends increasing the configuration of multi-functional facilities in the relevant area, such as building a comprehensive building integrating commerce, leisure, and residential services, to improve the functional mixing degree. At the same time, the functional adaptation is optimized according to the expected morphological conversion type of different spatial units. For example, for spatial units with an expected increase in building height, the supporting infrastructure is planned, such as strengthening the capacity of power, water supply and other facilities to adapt to the demand changes brought about by the new building height.

[0122] Through the processing of the multi-dimensional decision parameter set by the planning strategy generation model, the spatial form simulation results including the spatial control intensity level, the update time sequence priority, and the functional adaptation optimization plan are finally output. The spatial form simulation results provide specific planning guidance for urban planners, helping them formulate more in line with the development needs of the block planning scheme, and realizing the scientific planning and effective intervention of the spatial form of the historical and cultural block.

[0123] Further, after the step S120, the method further includes: Step S201: Perform integrity verification on the target spatio-temporal topological data set to detect missing data spatial units and abnormally fluctuating data points.

[0124] After generating the target spatio-temporal topological data set, perform integrity verification on it. For example, taking the spatial units in the block as an example, the target spatio-temporal topological data set contains various information of each spatial unit, such as spatial unit boundary data, building density change sequence, building height change sequence, functional zoning conversion sequence, activity intensity time series curve, and activity type distribution ratio, etc.

[0125] For the spatial unit boundary data, check whether there is missing boundary vertex coordinate information for a certain spatial unit. For example, in the boundary data of spatial unit K, there may be a situation where the coordinate of a certain vertex is not recorded, which belongs to missing data. In terms of the building density change sequence, check whether there is missing data for a certain time period. For example, in the building density change sequence of spatial unit L, there is no record during the period from 1980 to 1990, which is a manifestation of a missing data spatial unit.

[0126] For abnormally fluctuating data points, taking the building height change sequence as an example. Suppose the building height of spatial unit M has been growing steadily in the past few decades, such as growing 1-2 meters per year, but suddenly grows by 10 meters in a certain year, and this growth value deviates too much from the normal trend, which belongs to an abnormally fluctuating data point. In the activity intensity time series curve, abnormal situations may also occur. For example, the activity intensity of spatial unit N usually remains within a certain range on normal working days, but suddenly shows an extremely high activity intensity on a certain day, far exceeding the normal fluctuation range, which is an abnormally fluctuating data point in the activity intensity time series curve.

[0127] Step S202: Perform interpolation and filling processing on the detected missing data spatial units, and call the spatio-temporal topological data of adjacent spatial units for feature migration filling.

[0128] For the detected missing data spatial units, use the methods of interpolation and filling and feature migration filling for processing.

[0129] For other types of data, such as the missing vertex coordinates in the spatial unit boundary data, migration filling can also be performed based on the topological relationships and geometric shape features of adjacent spatial units. For example, the boundary shapes and topological adjacency relationships of adjacent spatial units have a certain degree of similarity. Based on the vertex coordinates and topological relationships of adjacent spatial units, the approximate positions of the missing vertex coordinates can be inferred. Suppose a vertex coordinate of an adjacent spatial unit has a parallel or similar geometric relationship with the edge where the missing vertex coordinate is located. Through this relationship, the possible values of the missing vertex coordinates can be calculated, thereby realizing the interpolation and completion of the spatial unit with missing data and the migration filling of features, making the target spatio-temporal topological data set more complete and accurate.

[0130] Step S203: Perform a smoothing correction process on the abnormal fluctuation data points, and replace the abnormal values beyond the set range based on the time series trend prediction value.

[0131] For the detected abnormal fluctuation data points, taking the abnormal increase in the building height of spatial unit M as an example, in this embodiment, the time series trend of its building height change is first analyzed. Through the analysis of the building height data over the past years, it is found that its growth trend generally conforms to a linear growth model (assumed), such as an increase of about 1.5 meters per year.

[0132] Set a reasonable fluctuation range. Suppose the normal fluctuation range is within ±20% of the average value. In a certain year, the building height of spatial unit M suddenly increased by 10 meters, which significantly exceeded the normal fluctuation range and was an abnormal value. In this embodiment, correction is performed based on the time series trend prediction value. According to the previous growth trend, the predicted growth value of the building height for that year should be the height of the previous year plus 1.5 meters. Suppose the building height of the previous year was 30 meters, then the predicted building height for that year should be 31.5 meters.

[0133] Replace the abnormal value of 10 meters beyond the set range with the predicted value of 31.5 meters to achieve the smoothing correction process of the abnormal fluctuation data points. For the abnormal fluctuation data points in the activity intensity time series curve, a similar method is also used. Analyze the time series trend of the activity intensity. Suppose it is found that the activity intensity on weekdays shows a certain periodicity and regularity. On a certain day, the activity intensity suddenly appears extremely high, exceeding the normal fluctuation range. By analyzing the activity intensity data in the same time period of past weekdays, a time series model (such as a moving average model, etc.) is established to predict the normal activity intensity value for that time period. Suppose the predicted value is 50 person-times per hour, while the actual abnormal value is 150 person-times per hour. Replace the abnormal value of 150 person-times per hour with the predicted value of 50 person-times per hour to make the activity intensity time series curve smoother and more reasonable, and reduce the interference of abnormal data on subsequent analysis and model training.

[0134] Step S204: Divide the verified target spatio-temporal topology data set into a training set and a validation set, optimize the parameters of the spatio-temporal evolution feature extraction model and the spatio-temporal evolution prediction model through the training set, and evaluate the accuracy of the spatial form simulation results through the validation set.

[0135] Divide the target spatio-temporal topology data set after integrity verification and data processing. According to a certain ratio (assuming 70% as the training set and 30% as the validation set), randomly select data of some spatial units to form the training set, and the remaining part forms the validation set.

[0136] For example, in a possible implementation manner, step S204 may include: Step S2041, extract features from the target spatio-temporal topology data set of each spatial unit in the training set to generate a training feature set, where the training feature set includes the static structure features, dynamic evolution pattern features, and spatial interaction dependence features of each spatial unit in the training set.

[0137] For example, for the spatial unit X in the training set, according to the previously described feature extraction method, extract its static structure features, dynamic evolution pattern features, and spatial interaction dependence features to generate a training feature set. This training feature set contains multi-faceted feature information of each spatial unit in the training set. Combine the above feature sets of all spatial units in the training set to form a complete training feature set.

[0138] Step S2042, input the training feature set into the spatio-temporal evolution prediction model to generate the predicted evolution probability distribution and the predicted evolution correlation strength of each spatial unit in the training set.

[0139] Step S2043, calculate the first error value between the predicted evolution probability distribution and the true evolution probability distribution of each spatial unit in the training set, and the second error value between the predicted evolution correlation strength and the true evolution correlation strength between adjacent spatial units in the training set.

[0140] In this embodiment, the first error value can be calculated by a conventional error calculation method (such as absolute error or mean square error, etc., assuming absolute error is used here), and the error calculation for the predicted evolution correlation strength and the true evolution correlation strength is the same.

[0141] Step S2044, construct a joint loss function according to the first error value and the second error value, and use the backpropagation algorithm to adjust the convolution kernel weights of the spatio-temporal evolution feature extraction model and the full connection layer parameters of the spatio-temporal evolution prediction model.

[0142] For example, assume that the combined loss function is the weighted sum of the two error values. Through this combined loss function, the convolutional kernel weights of the spatio-temporal evolution feature extraction model and the parameters of the fully connected layer of the spatio-temporal evolution prediction model are adjusted using the backpropagation algorithm. The backpropagation algorithm will backpropagate the error according to the value of the loss function and adjust the parameters in the model to gradually reduce the value of the loss function, thereby optimizing the performance of the model.

[0143] Step S2045: Input the target spatio-temporal topology data set of each spatial unit in the validation set into the spatio-temporal evolution feature extraction model after parameter tuning to generate a validation feature set.

[0144] For example, after feature extraction of the spatial unit Z in the validation set, its validation feature set is obtained, which includes static structure features, dynamic evolution pattern features, and spatial interaction dependence features.

[0145] Step S2046: Input the validation feature set into the spatio-temporal evolution prediction model after parameter tuning to generate a validation evolution probability distribution and a validation evolution correlation strength.

[0146] Specifically, the spatio-temporal evolution prediction model after parameter tuning has optimized and adjusted the model structure and parameters in order to more accurately analyze and predict the input data. For each spatial unit in the validation set, comprehensive analysis can be carried out based on the static structure features, dynamic evolution pattern features, and spatial interaction dependence features in its validation feature set. Finally, the validation evolution probability distribution of this spatial unit is generated, and at the same time, the validation evolution correlation strength between this spatial unit and adjacent spatial units is calculated. By performing such processing on all spatial units in the validation set, complete validation evolution probability distribution and validation evolution correlation strength data are obtained.

[0147] Step S2047: Extract the true evolution probability distribution of each spatial unit in the validation set and the true evolution correlation strength between adjacent spatial units, calculate the third error value between the validation evolution probability distribution and the true evolution probability distribution, and the fourth error value between the validation evolution correlation strength and the true evolution correlation strength.

[0148] First, accurately extract the true evolution probability distribution information of each spatial unit in the validation set from the pre-collected and organized real data records. These real data may come from long-term field observations, historical record analysis, and other reliable data sources, reflecting the true probability of different evolution states of spatial units in the actual development process.

[0149] At the same time, obtain the true evolution correlation strength between adjacent spatial units, which reflects the true degree of mutual influence and collaborative relationship between spatial units and surrounding units in the actual evolution process.

[0150] Next, for the third error value between the verified evolution probability distribution and the true evolution probability distribution, a suitable error calculation method is adopted, such as absolute error or mean square error, etc.

[0151] For the fourth error value between the verified evolution correlation strength and the true evolution correlation strength, it is also calculated according to the selected error calculation method.

[0152] The above error calculations are performed for each spatial unit in the verification set and its adjacent units to comprehensively evaluate the deviation degree between the model prediction results and the actual situation.

[0153] Step S2048: Generate an accuracy evaluation index based on the third error value and the fourth error value. When the accuracy evaluation index does not reach the preset threshold, increase the sample size of the training set and re - execute the parameter tuning step until the accuracy evaluation index reaches the preset threshold.

[0154] Specifically, the generation method of the accuracy evaluation index will be set according to the actual situation and requirements. For example, assume that the accuracy evaluation index is the reciprocal of the average of the two error values.

[0155] Compare the calculated accuracy evaluation index with the preset threshold. The preset threshold is a standard value set in advance according to factors such as project requirements, experience, and expectations for model accuracy.

[0156] When the accuracy evaluation index does not reach the preset threshold, it indicates that the prediction accuracy of the current model does not meet the requirements and needs to be further optimized. At this time, randomly select some additional spatial unit data from the original data set and add them to the training set to increase the sample size of the training set.

[0157] After increasing the sample size, re - execute the parameter tuning step until the accuracy evaluation index reaches the preset threshold. At this time, the prediction accuracy of the model meets the requirements, and the current spatio - temporal evolution feature extraction model and the spatio - temporal evolution prediction model are used as the final tuned models, which can more accurately predict and analyze the evolution of spatial units in the target block.

[0158] Figure 2 FIG. shows a schematic diagram of exemplary hardware and software components of a historical and cultural block spatial evolution simulation system 100 based on GIS spatio - temporal analysis that can implement the inventive concept provided by some embodiments of the present invention. For example, the processor 120 can be used on the historical and cultural block spatial evolution simulation system 100 based on GIS spatio - temporal analysis and is used to execute the functions in the present invention.

[0159] The historical and cultural block space evolution simulation system 100 based on GIS spatio-temporal analysis can be a general-purpose server or a special-purpose server, both of which can be used to implement the historical and cultural block space evolution simulation method based on GIS spatio-temporal analysis of the present invention.

[0160] For example, the historical and cultural block space evolution simulation system 100 based on GIS spatio-temporal analysis can include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and different forms of storage media 140, such as disks, ROM, or RAM, or any combination thereof. Exemplarily, the historical and cultural block space evolution simulation system 100 based on GIS spatio-temporal analysis can also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present invention can be implemented according to the above program instructions. The historical and cultural block space evolution simulation system 100 based on GIS spatio-temporal analysis also includes an I / O interface 150 between the computer and other input / output devices.

[0161] In addition, an embodiment of the present invention also provides a readable storage medium, in which computer-executable instructions are preset. When the processor executes the computer-executable instructions, the historical and cultural block space evolution simulation method based on GIS spatio-temporal analysis as described above is implemented.

[0162] It should be noted that, in order to simplify the description of the present invention disclosure and thus help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of the present invention, sometimes multiple features are merged into one embodiment, drawing, or description thereof.

Claims

1. A method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis, characterized in that, The method includes: Obtaining a target spatio-temporal data set of a target block, where the target spatio-temporal data set includes geospatial distribution data, building form change data, and human activity trajectory data in a historical period; Performing spatio-temporal topological structure preprocessing on the target spatio-temporal data set to generate a target spatio-temporal topological data set, where the target spatio-temporal topological data set contains a sequence of block spatial units with spatio-temporal associations and dynamic attribute labels of each spatial unit in the sequence of block spatial units; Based on a preset spatio-temporal evolution feature extraction model, performing feature extraction on the target spatio-temporal topological data set to obtain static structure features, dynamic evolution pattern features, and spatial interaction dependence features of each spatial unit; Invoking a spatio-temporal evolution prediction model to perform spatio-temporal evolution prediction on the static structure features, dynamic evolution pattern features, and spatial interaction dependence features to generate an evolution probability distribution of the spatial unit and an evolution association intensity between adjacent spatial units; Constructing a spatio-temporal evolution path map based on the evolution probability distribution and the evolution association intensity, and generating a spatial form simulation result and planning intervention decision parameters of the target block based on the spatio-temporal evolution path map.

2. The method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis according to claim 1, wherein The performing spatio-temporal topological structure preprocessing on the target spatio-temporal data set to generate a target spatio-temporal topological data set includes: Performing coordinate calibration processing on the geospatial distribution data to obtain spatial unit boundary data in a unified coordinate system, where the spatial unit boundary data contains geometric shape vertex coordinates and topological adjacency relationships of each spatial unit; Performing time series alignment processing on the building form change data to obtain a building density change sequence, a building height change sequence, and a functional zoning conversion sequence that match the spatial unit boundary data; Performing spatial rasterization processing on the human activity trajectory data to generate a human activity heat distribution map corresponding to the spatial unit boundary data, and extracting the activity intensity time series curve and activity type distribution ratio of each spatial unit in the human activity heat distribution map; Performing spatio-temporal topological association coding on the spatial unit boundary data, the building density change sequence, the building height change sequence, the functional zoning conversion sequence, the activity intensity time series curve, and the activity type distribution ratio to generate the target spatio-temporal topological data set, where the dynamic attribute label of each spatial unit includes a building density change rate, a function conversion frequency, and an activity intensity fluctuation amplitude.

3. The method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis according to claim 1, wherein The based on a preset spatio-temporal evolution feature extraction model, performing feature extraction on the target spatio-temporal topological data set to obtain static structure features, dynamic evolution pattern features, and spatial interaction dependence features of each spatial unit includes: Extracting the spatial unit boundary data in the target spatio-temporal topological data set, and invoking a spatial form encoder to generate an initial structure feature vector of each spatial unit, where the initial structure feature vector includes shape complexity, area change trend, and the number of adjacent units; Perform temporal convolution processing on the building density change sequence, building height change sequence, and functional zoning conversion sequence in the target spatio-temporal topological data set to generate a first dynamic evolution feature vector, where the first dynamic evolution feature vector includes the building density fluctuation period, height growth gradient, and functional conversion trigger threshold; Perform spatio-temporal attention processing on the activity intensity time series curve and activity type distribution ratio in the target spatio-temporal topological data set to generate a second dynamic evolution feature vector, where the second dynamic evolution feature vector includes the activity intensity peak phase, type mixing degree, and spatial diffusion direction; Perform cross-scale feature splicing on the initial structure feature vector, the first dynamic evolution feature vector, and the second dynamic evolution feature vector to obtain a static structure feature; Perform graph convolution processing on the topological adjacency relationship in the target spatio-temporal topological data set to generate a spatial interaction dependence feature, where the spatial interaction dependence feature includes the density conduction coefficient between adjacent units, the functional compatibility index, and the activity diffusion resistance value; Perform dynamic weighted correction on the static structure feature based on the dynamic attribute labels in the target spatio-temporal topological data set to obtain a dynamic evolution pattern feature.

4. The method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis according to claim 3, characterized in that The performing graph convolution processing on the topological adjacency relationship in the target spatio-temporal topological data set to generate a spatial interaction dependence feature includes: Construct an undirected graph structure with the spatial units in the spatial unit boundary data as nodes and the topological adjacency relationship as edges; Initialize the initial feature vector of each node in the undirected graph structure, where the initial feature vector is generated by splicing the shape complexity, area change trend, and number of adjacent units in the static structure feature with the building density fluctuation period and functional conversion trigger threshold in the dynamic evolution pattern feature; Perform multi-layer graph convolution network processing on the undirected graph structure, and layer by layer aggregate the adjacent node feature vectors of each node to generate an intermediate spatial interaction feature; After each layer of graph convolution network processing, input the intermediate spatial interaction feature into a gated recurrent unit, generate a neighborhood feature fusion weight based on the functional conversion frequency in the dynamic attribute labels, and perform weighted superposition on the intermediate spatial interaction feature and the current node feature vector to update the node feature vector to obtain the node feature vector output by the final layer; Perform spatial scale alignment processing on the node feature vector output by the final layer, and perform spatial position encoding on the feature vector of each node with the geometric shape vertex coordinates of the spatial unit boundary data to generate a spatial interaction dependence feature including the density conduction gradient, functional compatibility matrix, and activity diffusion direction vector.

5. The method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis according to claim 1, wherein The calling the spatio-temporal evolution prediction model to perform spatio-temporal evolution prediction on the static structure feature, dynamic evolution pattern feature, and spatial interaction dependence feature to generate the evolution probability distribution of the spatial unit and the evolution correlation strength between adjacent spatial units includes: Input the static structure feature into the first branch network of the spatio-temporal weight allocation model to generate an internal evolution weight coefficient for the spatial unit, where the internal evolution weight coefficient reflects the influence degree of the historical form on the current evolution direction; Input the dynamic evolution pattern features into the second branch network of the spatio-temporal weight allocation model to generate a time decay adjustment coefficient, which reflects the contribution decay rate of data in different periods to the prediction result; Input the spatial interaction dependence features into the third branch network of the spatio-temporal weight allocation model to generate a neighborhood coupling strength coefficient, which reflects the cooperative or competitive relationship of the evolution behaviors of adjacent spatial units; Construct a spatio-temporal joint weight matrix according to the internal evolution weight coefficient, the time decay adjustment coefficient and the neighborhood coupling strength coefficient, and perform weighted fusion on the static structure features, the dynamic evolution pattern features and the spatial interaction dependence features to generate a spatio-temporal fusion feature vector; Input the spatio-temporal fusion feature vector into the evolution probability prediction layer to generate the evolution probability distribution of the spatial unit; Input the spatio-temporal fusion feature vector into the correlation strength prediction layer to generate the evolution correlation strength between adjacent spatial units.

6. The method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis according to claim 1, wherein The construction of the spatio-temporal evolution path map according to the evolution probability distribution and the evolution correlation strength includes: Construct an initial spatio-temporal evolution graph structure with each spatial unit as a node and the evolution correlation strength as the edge weight; Update the node states of the initial spatio-temporal evolution graph structure according to the evolution probability distribution to generate a spatio-temporal evolution graph with probability weights; Perform path search processing on the spatio-temporal evolution graph with probability weights, and extract the edges with probability weights greater than the set weight to form a set of main evolution paths; Perform loop detection on the set of main evolution paths. If a closed loop is detected, decay and correct the evolution probability within the loop according to the time decay adjustment coefficient; Overlay and render the set of main evolution paths after decay correction with the spatial unit boundary data to generate the spatio-temporal evolution path map, which includes a historical form preservation area, a high-probability evolution area and a neighborhood cooperative evolution corridor.

7. The method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis according to claim 1, characterized in that The generation of the spatial form simulation result and the planning intervention decision parameters of the target block based on the spatio-temporal evolution path map includes: Extract the set of spatial units in the historical form preservation area of the spatio-temporal evolution path map, calculate the variance value of the building density change sequence and the mean value of the function conversion frequency, and generate a form stability index; Extract the set of spatial units in the high-probability evolution area of the spatio-temporal evolution path map, and generate the maximum probability evolution direction and the expected form conversion type according to its evolution probability distribution; Extract the edge weight distribution of the neighborhood cooperative evolution corridor in the spatio-temporal evolution path map to generate a corridor width adjustment parameter and a function mixing degree improvement coefficient; Construct a multi-dimensional decision parameter set according to the form stability index, the maximum probability evolution direction, the expected form conversion type, the corridor width adjustment parameter and the function mixing degree improvement coefficient; Input the multi-dimensional decision parameter set into the planning strategy generation model, and output a spatial form simulation result including the spatial control intensity level, the update time sequence priority and the function adaptation optimization plan.

8. The method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis according to claim 7, characterized in that Inputting the multi-dimensional decision parameter set into the planning strategy generation model, and outputting the spatial form simulation results including the spatial control intensity level, the update time sequence priority, and the function adaptation optimization plan, including: Classifying the form stability index, and activating the form protection strategy when the form stability index is lower than the stability threshold, to generate the building height limit parameter and the material use constraint condition; Detecting conflicts in the maximum probability evolution direction, and inserting a direction correction factor and recalculating the evolution probability distribution when it is detected that the maximum probability evolution direction conflicts with the historical context protection requirements; Evaluating the carrying capacity of the expected form conversion type, and generating a list of infrastructure expansion requirements and a traffic organization optimization plan when the evaluation result indicates that the carrying capacity exceeds the preset capacity; Calculating the penetration buffer distance between adjacent functional areas according to the corridor width adjustment parameter, and generating an interface transition processing strategy; Adjusting the land use compatibility rule based on the function mixing degree improvement coefficient, and generating the mixed function ratio control parameter and the public space configuration standard; Sorting the form protection strategy, the direction correction factor, the infrastructure expansion requirement list, the traffic organization optimization plan, the interface transition processing strategy, and the mixed function ratio control parameter in terms of strategy priority to generate the planning intervention decision parameter; 9. The method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis according to claim 1, wherein After preprocessing the spatio-temporal topology structure of the target spatio-temporal data set to generate the target spatio-temporal topology data set, the method further includes: Performing integrity verification on the target spatio-temporal topology data set to detect missing data spatial units and abnormally fluctuating data points; Performing interpolation and completion processing on the detected missing data spatial units, and calling the spatio-temporal topology data of adjacent spatial units for feature migration filling; Performing smoothing and correction processing on the abnormally fluctuating data points, and replacing the abnormal values outside the set range based on the time series trend prediction value; Dividing the verified target spatio-temporal topology data set into a training set and a validation set, and tuning the parameters of the spatio-temporal evolution feature extraction model and the spatio-temporal evolution prediction model through the training set, and evaluating the accuracy of the spatial form simulation results through the validation set; 10. A spatial evolution simulation system of historical and cultural blocks based on GIS spatio-temporal analysis, characterized in that, Including a processor and a memory, the memory is connected to the processor, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the method for simulating the spatial evolution of historical and cultural blocks based on GIS spatio-temporal analysis described in any one of claims 1-9 above.

Images