A city renewal planning system and method based on big data analysis

Abstract

The application discloses a kind of urban renewal planning system and method based on big data analysis, belong to urban renewal planning technical field.The method includes: extracting city land use type transition trajectory and infrastructure load fluctuation sequence, constructs city space resilience coefficient matrix and locates the block to be updated;Obtain the subdivision atlas of property block and the distribution of volume rate, generate block integration potential index, generate function replacement demand vector in combination with population flow time series tensor and calculate development intensity adjustment coefficient;According to the adjustment coefficient and the resilience matrix, generate three-dimensional space simulation contour, obtain constructability verification results by underground pipeline collision detection, adjust block sequence and output urban renewal planning scheme.The scheme can accurately identify the resilience recession area, improve the consistency of block integration feasibility and functional demand response.

Description

Technical Field

[0001] This invention relates to the field of urban renewal planning technology, specifically to an urban renewal planning system and method based on big data analysis. Background Technology

[0002] In urban renewal planning, the identification of areas to be renewed has long relied on static, single indicators such as building quality, construction age, or aging pipelines. This approach fails to consider the rapid evolution of urban land use over time and the dynamic fluctuations in infrastructure carrying capacity, making it difficult to reveal the true resilience of neighborhoods in the face of population agglomeration and business transformation. Furthermore, planning indicators are often assigned universally across neighborhoods, lacking quantitative descriptions of the fragmented ownership of land parcels and the differences in current floor area ratios. The difficulty of land consolidation cannot be objectively assessed in advance, leading to frequent ownership coordination dilemmas during the actual implementation of renewal plans, resulting in a misalignment between planning indicators and feasibility. In addition, the functional allocation of neighborhoods is often based on established higher-level plans, lacking in-depth analysis of dynamic activity characteristics such as real-time population occupancy and commuting flow. This leads to a discrepancy between the completed functional layout and the actual needs of citizens, resulting in prominent issues such as vacant commercial spaces or insufficient residential service facilities. Collision checks on underground pipelines introduced later in the planning process often only reveal conflicts during the construction phase, causing large-scale pipeline relocation and project delays. How to construct a spatial resilience metric by leveraging land use change trajectories and infrastructure load fluctuation sequences, accurately delineate the scope of blocks to be renewed, and use resilience levels as a rigid constraint for three-dimensional spatial construction planning is a problem that current technologies have not yet solved. Another technical challenge in the refinement of urban renewal is how to extract the potential for land consolidation from property rights subdivision maps and how to mutually drive functional replacement demands derived from population flow temporal tensors, thereby guiding adjustments to development intensity. Summary of the Invention

[0003] The purpose of this invention is to provide an urban renewal planning system and method based on big data analysis. It addresses the problems in existing technologies, such as the lack of quantitative basis for resilience in the selection of street regeneration sites, the lack of assessment of the potential for land integration, and the disconnect between functional needs and dynamic population activities. By integrating multi-source urban big data, it realizes a closed-loop planning from resilience assessment to land integration, functional replacement, and spatial verification, thereby improving the scientific nature and implementation efficiency of the renewal plan.

[0004] To achieve the above objectives, this invention provides the following technical solution: This invention offers an urban renewal planning method based on big data analysis, which integrates multi-source urban data to achieve end-to-end planning decisions from current status diagnosis to scheme generation. This method extracts urban land use type change trajectories and infrastructure load fluctuation sequences from time-series data of the urban geographic information system, thereby reconstructing the evolutionary patterns of urban space and the dynamic changes in infrastructure carrying capacity. An urban spatial resilience coefficient matrix is ​​constructed using the land use type change trajectories and infrastructure load fluctuation sequences, and based on this matrix, a set of neighborhoods with insufficient spatial resilience and urgent need for renewal is automatically located, thus establishing the identification of renewal needs based on dual evidence of neighborhood decline trends and infrastructure pressure.

[0005] Preferably, after obtaining the set of blocks to be updated, the detailed property rights map of each block is retrieved, and combined with the current plot ratio distribution, a land consolidation potential index is generated, reflecting the ease of land ownership consolidation and the dispersion level of development intensity. Based on this index, the set of blocks to be updated is hierarchically screened, and blocks with the strongest consolidation potential are prioritized for inclusion in the sequence of blocks to be updated, thereby improving the operability and economies of scale of subsequent updates. Furthermore, real-time population heat map distribution and public transportation card swipe records corresponding to the sequence of blocks to be updated are collected to construct a population flow time series tensor containing population residence characteristics and travel patterns. Based on this time series tensor and the land consolidation potential index, a functional replacement demand vector for each block to be updated is generated, so that functional adjustment decisions match the actual rhythm of population activity.

[0006] As a technical solution of this invention, based on the functional replacement demand vector and the current control detailed planning indicators, the development intensity adjustment coefficient of each initiated renewal block is calculated, realizing the transmission from macro-planning control to micro-plot intensity assignment. The development intensity adjustment coefficient is combined with the urban spatial resilience coefficient matrix to generate a three-dimensional spatial construction simulation outline of each initiated renewal block, ensuring that the building volume layout responds to renewal needs while being constrained by spatial resilience, avoiding further degradation of resilience due to overdevelopment. Subsequently, the three-dimensional spatial construction simulation outline is subjected to collision detection with the underground pipeline integrated database to obtain constructability verification results. Based on these results, the sequence of initiated renewal blocks is adjusted, ultimately outputting an urban renewal planning scheme compatible with underground space conditions, avoiding construction conflicts from the outset.

[0007] Corresponding to the above method, the present invention also provides an urban renewal planning system based on big data analysis, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the aforementioned urban renewal planning method. This system can automatically execute the entire process from data extraction, resilience matrix construction, potential index calculation, block selection, functional requirement generation, development intensity estimation to three-dimensional contour simulation and pipeline collision detection, thereby significantly improving planning efficiency while ensuring the scientific nature of the planning.

[0008] The technical effects and advantages provided by the present invention in the above technical solution are as follows: By dividing the urban area into spatial grids, the land use type change trajectory of each grid is encoded as a change vector. The first deviation between the change vector and the preset stable land use pattern vector is calculated. Simultaneously, the infrastructure load fluctuation sequence is encoded as a load vector, and the second deviation between the load vector and its historical mean vector is calculated. The first and second deviations are weighted and summed to obtain the spatial resilience coefficient of each grid, and arranged to form an urban spatial resilience coefficient matrix. The blocks containing grids with spatial resilience coefficients lower than a preset resilience threshold in this matrix are marked as the set of blocks to be updated. When generating the three-dimensional spatial construction simulation outline of each block to be updated, after obtaining the preliminary building blocks based on the development intensity adjustment coefficient and the existing building foundation and height, the spatial resilience coefficient of the corresponding grid in the urban spatial resilience coefficient matrix is ​​used as an elevation constraint to reduce and correct the top elevation of the preliminary building blocks. This process enables a dynamic match between the renewal construction intensity and the resilience level of the blocks, avoiding high-intensity development in areas with fragile infrastructure or drastic fluctuations in land use functions, reducing the risk of overloaded operation of public facilities and safety hazards, and realizing the automatic generation of building forms under resilience constraints. By retrieving the ownership boundary vector data and current floor area ratio distribution of property parcels within the block to be updated, a detailed property parcel map is generated. The number of adjacent property parcel pairs with the same ownership entity within each block is counted. Simultaneously, the variance of the floor area ratio distribution of all property parcels within the block is calculated. After normalizing the quantity and variance, the summation yields the land parcel integration potential index. Based on the land parcel integration potential index, blocks are hierarchically screened and sorted to determine the sequence of blocks to initiate updates. Using a population flow time-series tensor constructed from mobile signaling and public transportation card swipe data, the daytime and nighttime population occupancy ratios of each block are extracted. The difference between the daytime and nighttime occupancy ratios is multiplied by the land parcel integration potential index to obtain the commercial and residential function offset. This offset is then superimposed with the current planning's commercial and residential land area ratio to obtain the functional replacement demand vector, and the development intensity adjustment coefficient is calculated accordingly. Therefore, by incorporating the ease or difficulty of integrating land ownership with the actual needs of residents for commuting and staying, the decision on the timing and functional allocation of the renewal is made. This reduces obstacles to expropriation negotiations caused by fragmented property rights, ensures that the functions of the renewed neighborhood are highly compatible with the daytime and nighttime activity characteristics of citizens, alleviates the contradiction between work and residence, and improves the efficiency of public service facilities. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0010] Figure 1 This is a flowchart of urban renewal planning methods based on big data analysis; Figure 2This is a flowchart illustrating the method for generating a simulated outline for the construction of a 3D street block. Figure 3 This is a tiered map of the potential for consolidation of blocks in the area awaiting renovation. Figure 4 This is a sorting chart of the tendency values ​​for initiating and updating the street function replacement. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] See Figure 1 This invention provides an urban renewal planning method based on big data analysis, comprising: extracting urban land use type change trajectories and infrastructure load fluctuation sequences from time-series data of an urban geographic information system; constructing an urban spatial resilience coefficient matrix based on the land use type change trajectories and infrastructure load fluctuation sequences, and locating a set of blocks to be renewed based on the urban spatial resilience coefficient matrix; obtaining a property rights subdivision map of each block in the set of blocks to be renewed, and generating a land consolidation potential index based on the property rights subdivision map and the current plot ratio distribution; performing hierarchical screening of the set of blocks to be renewed based on the land consolidation potential index to determine a sequence of blocks to be renewed; and collecting the sequence of blocks to be renewed. Based on real-time population heat map distribution and public transportation card swipe records, a population flow time series tensor is constructed. According to the population flow time series tensor and the land parcel integration potential index, a functional replacement demand vector for each initiated urban renewal block is generated. Based on the functional replacement demand vector and current control detailed planning indicators, a development intensity adjustment coefficient for each initiated urban renewal block is calculated. Based on the development intensity adjustment coefficient and the urban spatial resilience coefficient matrix, a three-dimensional spatial construction simulation outline for each initiated urban renewal block is generated. Collision detection is performed between the three-dimensional spatial construction simulation outline and the underground pipeline integrated database to obtain constructability verification results. Based on the constructability verification results, the sequence of initiated urban renewal blocks is adjusted, and an urban renewal planning scheme is output.

[0013] Example 1: In practice, land use classification raster data from the urban geographic information system at multiple historical time segments are acquired. These historical time segments include the first time segment, the second time segment, and up to the Tth time segment, where T is an integer greater than or equal to 3. Each time segment's land use classification raster data contains land use type codes at the planar location, in pixels. These codes adopt the category codes specified in the national standard "Classification of Urban Land Use and Standards for Planning and Construction Land," for example, residential land is coded as R, commercial and service facilities land as B, and industrial land as M. The land use classification raster data from all time segments are overlaid, ensuring that pixels at different time segments within the same geographic unit are spatially strictly aligned. After overlay, the land use type change chain for each geographic unit is extracted. This chain records the land use type code sequence for that geographic unit from the first time segment to the Tth time segment, with changes in the codes reflecting the trajectory of land use type changes.

[0014] Infrastructure operation records from the city's geographic information system (GIS) at multiple historical time segments are acquired, with each historical time segment corresponding one-to-one with the time segments of the land use classification raster data. These infrastructure operation records include those for water supply network nodes, drainage network nodes, power line nodes, and communication base station nodes. For water supply network nodes, the node flow load values ​​for each node from the first time segment to the Tth time segment are extracted. The node flow load value is the ratio of the average flow monitored by the node within the corresponding time segment to the design flow. For drainage network nodes, the node fill load values ​​for each drainage network node from the first time segment to the Tth time segment are extracted. The node fill load value is the ratio of the water depth in the pipe to the pipe diameter. For power line nodes, the line load rate values ​​for each power line node from the first time segment to the Tth time segment are extracted. The line load rate value is the ratio of the actual transmitted power to the rated transmitted power. For each communication base station node, the channel occupancy rate is extracted from the first time segment to the Tth time segment. The channel occupancy rate is the ratio of the average number of channels occupied during busy hours to the total number of channels. The load values ​​of all the above infrastructure nodes are aggregated by node and arranged in chronological order to obtain the infrastructure load fluctuation sequence for each infrastructure node. An infrastructure load fluctuation sequence consists of T sequentially arranged load values.

[0015] In practice, the urban area is divided into equal-sized spatial grids, each square with a side length equal to the pixel resolution of the land use classification raster data to maintain consistent spatial granularity. For the spatial grid in the i-th row and j-th column, the land use type change trajectory on the corresponding geographic unit is encoded as a change vector. The dimension of the change vector is equal to the total number of time segments T. The t-th dimension of the change vector is a numerical value mapped from the land use type code. The mapping method assigns discrete values ​​according to the alphabetical order of the category codes or functional similarity. For example, residential land is mapped to the value 1, commercial and service facilities land is mapped to the value 2, industrial land is mapped to the value 3, and so on. The values ​​do not have any order of magnitude; they are only used to distinguish land use types.

[0016] The load fluctuation sequence of the infrastructure nodes contained in the spatial grid in the i-th row and j-th column is encoded into a load vector. If there are multiple infrastructure nodes in a spatial grid, the average load value of the multiple infrastructure nodes at each time segment is taken as the sub-vector of the load vector at the corresponding time segment. The load vector has dimension T, and the t-th dimension sub-vector of the load vector is the average load value of the spatial grid in the t-th time segment.

[0017] In some embodiments, a first deviation is calculated between the transition vector of each spatial grid and a preset stable land use pattern vector. The stable land use pattern vector is a preset reference vector whose components in all T dimensions take the same land use type code value, representing the ideal stable state where the land use type of the spatial grid has never changed throughout all historical time segments. The first deviation is calculated using the square of the Euclidean distance; that is, for the spatial grid in the i-th row and j-th column, the first deviation... It is equal to the sum of the squares of the differences between the spatial grid transition vector and the stable land use pattern vector in each time dimension. In another implementation, the first deviation is calculated using cosine distance. The first deviation is obtained by calculating the cosine of the angle between the transition vector and the stable land use pattern vector, and then subtracting the cosine from 1.

[0018] Calculate the second deviation between the load vector of each spatial grid and its historical mean vector. The historical mean vector is obtained by averaging the load values ​​of all spatial grids for each time segment. The historical mean vector has dimension T, and its t-th dimension is the arithmetic mean of the load values ​​of all spatial grids at time segment t. The second deviation is calculated using the mean squared error; that is, for the spatial grid in row i and column j, the second deviation... It equals the sum of the squares of the differences between the corresponding dimension vectors of the load vector and the historical mean vector, divided by the dimension T. In another implementation, the second deviation is calculated using the mean absolute error.

[0019] The spatial toughness coefficient of each spatial grid is obtained by weighted summing of the first and second deviations. in, This represents the spatial toughness coefficient of the spatial grid located in the i-th row and j-th column. This represents the first deviation of the spatial grid in the i-th row and j-th column. This represents the second deviation of the spatial grid in the i-th row and j-th column. The fusion weight has a value range of 0 to 1, and the value is set to 0.55. The 0.55 value is determined by collecting historical spatial grid data from fifteen urban renewal zones, extracting land use change trajectories and infrastructure load fluctuation data before renewal in each zone, calculating the first and second deviations, and using manually labeled renewal necessity levels as a reference. Logistic regression is employed to fit the data, maximizing the goodness of fit between the spatial resilience coefficient and renewal necessity. The optimized value is 0.55.

[0020] The spatial resilience coefficients of all spatial grids are arranged according to their row and column positions, forming a city spatial resilience coefficient matrix. The matrix dimension is the number of rows M multiplied by the number of columns N, where M is the total number of rows and N is the total number of columns. Spatial grids with spatial resilience coefficients lower than a preset resilience threshold are extracted from the city spatial resilience coefficient matrix. The resilience threshold is set at 0.35, based on the lower quartile values ​​of the spatial resilience coefficient distribution corresponding to stable areas that have not undergone major land use and infrastructure renovations in the past three planning cycles. This threshold has been verified to effectively distinguish between stable and degraded areas. The blocks containing the extracted spatial grids are marked. Each block is a spatial unit defined by administrative or natural boundaries. If a block contains any marked spatial grid, it is included in the block set to be updated. The block set to be updated consists of multiple blocks.

[0021] Example 2: In practice, the cadastral database records for each block in the set of blocks to be updated are retrieved. These records are stored in the real estate registration system of the natural resources management department and include parcel graphic data, ownership attribute data, and planning permit data. The ownership boundary vector data of all property parcels within each block is analyzed. This data consists of closed polygons formed by concatenating the coordinates of boundary points determined by land surveys, recording the spatial location and geometric boundaries of each property parcel. All property parcel ownership boundary vector data are then cropped and combined according to the block boundaries to generate a detailed property parcel map. This detailed map is a layered vector map containing all property parcel polygons within the block and their unique parcel identifiers.

[0022] The boundary adjacency relationships and ownership information between adjacent property parcels are extracted from the detailed property parcel map. Boundary adjacency relationships are obtained through spatial topology calculation. The spatial topology calculation process involves performing face-to-face adjacency analysis on all polygons of the property parcels to determine whether two polygons share a common boundary with a length greater than 0.05 meters. If they share a common boundary and its length exceeds the aforementioned threshold, a boundary adjacency relationship is determined between the two property parcels. Ownership information is extracted from the rights holder field in the cadastral database, which records the name or unified social credit code of the land user of the property parcel. This ownership information is then associated and stored with each property parcel. The current floor area ratio (FAR) for each property parcel is obtained. The current FAR is read from the planning permit data recorded in the cadastral database, and its value equals the total approved building area of ​​the property parcel divided by the land area of ​​the property parcel.

[0023] Based on boundary adjacency relationships, count the number of adjacent property parcel pairs with the same ownership entity within each block. Traverse all identified adjacent property parcel pairs within a block, checking if the ownership entity information of the two parcels in the pair is identical. If the land user's name or unified social credit code of the two parcels is exactly the same, then count the adjacent property parcel pair as a pair with the same ownership entity. Accumulate the number of all property parcel pairs within the block that meet the criteria to obtain the total number of adjacent property parcel pairs with the same ownership entity in that block.

[0024] Calculate the variance of the floor area ratio distribution for all property parcels within each block. Obtain the current floor area ratio values ​​for all property parcels within the block, forming a numerical sequence of size m, where m is the total number of property parcels within the block. Calculate the arithmetic mean of the numerical sequence, then calculate the square of the difference between each current floor area ratio value and the arithmetic mean. Sum all the squared values ​​and divide by m to obtain the variance of the floor area ratio distribution.

[0025] The number of adjacent land parcels with the same ownership entity obtained from the above statistics and the variance of the calculated plot ratio distribution are normalized. The normalization process uses a maximum-minimum normalization method. For all blocks in the set of blocks to be updated, the maximum and minimum values ​​of the number of adjacent land parcels with the same ownership entity, as well as the maximum and minimum values ​​of the variance of the plot ratio distribution for all blocks, are obtained. The land consolidation potential index for each block is calculated using the following formula: in, This represents the land consolidation potential index for the k-th block; This represents the number of adjacent property plots with the same ownership in the k-th block; This represents the minimum number of adjacent property rights pairs with the same ownership entity across all blocks in the block set to be updated; This represents the maximum number of adjacent property rights pairs with the same ownership entity across all blocks in the block set to be updated; This represents the variance of the floor area ratio distribution in the k-th block; This represents the minimum variance of the floor area ratio distribution of all blocks in the block set to be updated; This represents the maximum variance of the floor area ratio distribution of all blocks in the block set to be updated.

[0026] In practice, the blocks in the set to be updated are initially sorted according to their land consolidation potential index from largest to smallest, resulting in an ordered list. Blocks ranked higher in the list correspond to higher land consolidation potential indices. A tiered interval step size is set based on the total number of blocks in the set. The total number of blocks is denoted as Q, and the tiered interval step size is set to a maximum integer not exceeding Q / 4, ensuring that all blocks are evenly divided into no more than four potential tiers. Following the initial sorting, starting from the first block in the ordered list, blocks are sequentially assigned to the same potential tier for every consecutive tiered interval step size, until all blocks are assigned. If any remaining blocks are less than one tiered interval step size, they are assigned to the last potential tier.

[0027] Calculate the average land consolidation potential index for all blocks within each potential tier. The average land consolidation potential index is the arithmetic mean of the land consolidation potential indices of all blocks within that potential tier. Compare the average land consolidation potential indices of each potential tier, extract the potential tier with the highest average land consolidation potential index, and mark all blocks in the potential tier with the highest average land consolidation potential index as the first batch of candidate blocks.

[0028] The spatial distribution coordinates of the first batch of candidate blocks are overlaid with the existing key urban development axes. The existing key urban development axes are derived from urban spatial structure elements determined in the city's master plan or territorial spatial master plan, and are stored in the form of polylines or buffer polygons. The overlay analysis uses GIS spatial overlay operations to specifically determine whether the boundary polygons of each first batch of candidate blocks intersect with the buffer polygons of the key development axes. If the boundary polygons of a batch of candidate blocks intersect with the buffer polygons of the key development axes, then that candidate block is selected to enter the sequence of blocks to be updated. The sequence of blocks to be updated is an ordered set of blocks arranged from high to low according to their land consolidation potential index.

[0029] See Figure 3 In the graph, the horizontal axis represents the block number and the vertical axis represents the land consolidation potential index. The curve shows the trend of blocks being sorted from largest to smallest according to their land consolidation potential index. The four colors in the figure represent the potential tiers: red for the first potential tier, blue for the second, green for the third, and purple for the fourth. The number of blocks in each potential tier is approximately 150, 150, 150, and 150 respectively. Overall, the set of blocks to be updated is evenly divided into no more than four tiers, conforming to the tiered layering requirement in Example 2.

[0030] The first potential tier (red) of blocks has a land consolidation potential index ranging from approximately 0.42 to 0.9. The overall index value is relatively high and shows a gradually decreasing trend, indicating that this tier of blocks performs well in terms of the sum of the normalized variances of the number and floor area ratio distributions of adjacent property plots with the same ownership, demonstrating strong land consolidation potential. The second potential tier (blue) has an index ranging from approximately 0.32 to 0.5. The index value is generally lower than the first tier and also shows a gradually decreasing trend. The third potential tier (green) has an index value concentrated between 0.2 and 0.40, with a continuously declining trend. The fourth potential tier (purple) has the lowest index value, ranging from 0 to 0.30, indicating weak land consolidation potential in this tier of blocks.

[0031] Example 3: In practice, population heat map grids for each block within the updated block sequence are obtained from mobile operator signaling data, covering consecutive weekdays and weekends. The mobile operator signaling data includes anonymized user identifiers, signaling trigger timestamps, and serving base station cell codes, with the signaling trigger timestamps accurate to the second. The base station cell codes are associated with base station geographic location information, mapping each signaling record to a regular grid covering the urban area, with each grid cell being a square grid with sides of 250 meters. Consecutive weekdays are defined as five consecutive days from Monday to Friday, and weekends are defined as the following two consecutive days, Saturday and Sunday, excluding public holidays. The number of different anonymized user identifiers appearing in each grid is aggregated according to an hourly time window, yielding the hourly population heat map value for each grid. The hourly population heat map values ​​for all grids on weekdays and weekends are concatenated to form a population heat map data sequence covering all seven consecutive days and a total of 168 time slices. For each initiated update block, population heatmap values ​​of all rasters within the block boundary vector are extracted. The population heatmap values ​​of each raster are then arranged chronologically to generate a population heatmap time-series matrix for that block. The rows of the population heatmap time-series matrix correspond to the spatial location indices of the rasters within the block, the columns correspond to the 168 sequential time slice indices, and the matrix elements are the population heatmap values ​​of the rasters within their corresponding time slices.

[0032] The system retrieves entry and exit card swipe records for bus and rail stations surrounding the initiated update block sequence from the public transportation automated fare collection system. The automated fare collection system includes bus fare terminals and rail transit automated fare gates, recording the timestamp, station number, and transaction type for each entry and exit transaction. A buffer zone with a radius of 500 meters is established based on the boundary polygon of the initiated update block; bus and rail stations falling within this buffer zone are identified as surrounding stations. For each surrounding station, the number of entry and exit card swipes is counted within a one-hour time window, identical to the population heat map data. The sum of these two counts yields the passenger flow value for each time slot. The passenger flow values ​​for each station are arranged sequentially by time slot to form a 168-dimensional passenger flow time-series vector.

[0033] The population heatmap time-series matrix of all grids within the same starting update block is concatenated with the passenger flow time-series vectors of all surrounding stations along the time dimension to form the population flow matrix within the block. The concatenation process is as follows: For a population heatmap time-series matrix with g rows and t columns, and s stations, a new matrix with g+s rows and t columns is constructed; the first g rows of the new matrix contain the element values ​​of each row of the population heatmap time-series matrix, and the last s rows contain the transpose of the station passenger flow time-series vectors, with each station occupying one row and each column corresponding to the same time slice. The population flow matrix within the block completely represents the population distribution and flow status of residential and travel space units within a single block over continuous time.

[0034] The population flow matrices of all initiated update blocks are stacked along the block dimension to construct a population flow time series tensor. The population flow time series tensor is stored in a three-dimensional array. The first dimension index is the number of the initiated update block in the sequence, the second dimension index is the spatial unit number within the block, and the spatial unit includes grids and stations. The third dimension index is the time slice number.

[0035] In some embodiments, the population flow time series tensor undergoes a periodic decomposition along the time dimension, employing a direct statistical method based on time segmentation. A 24-hour day is divided into daytime, nighttime, and transitional periods. The daytime period is defined as 08:00 to 18:00 daily, the nighttime period as 20:00 to 06:00 the following day, and the transitional periods as 06:01 to 07:59 and 18:01 to 19:59. Population data during the transitional periods are not included in subsequent calculations. For each starting update block in the population flow time series tensor, the population or passenger flow values ​​for each time slot during the daytime period across all weekdays and weekends for all spatial units of that block are extracted. The total population of the block for each time slot is obtained by summing the data by spatial unit and time slot. Then, the arithmetic mean of the total population of the block across all daytime time slots is calculated to obtain the daytime resident population of that block. Similarly, the total population of the block across all nighttime time slots is extracted and its arithmetic mean is calculated to obtain the nighttime resident population of that block. The difference ratio between daytime and nighttime occupancy is calculated by dividing the difference ratio by the sum of the daytime and nighttime occupancy. This difference ratio is then multiplied by the land consolidation potential index of the initiated renewal block to obtain the commercial and residential function shift of that block. The land consolidation potential index is derived from the aforementioned generation step and is redefined for the u-th initiated renewal block. , The range of values ​​is determined by the normalized summation result, and it remains positive.

[0036] Obtain the area ratio of commercial land to residential land in the current detailed control plan of the block where the update is initiated. This is equal to the total planned commercial land area within the block divided by the total planned residential land area. The offset between commercial and residential functions is then added to the area ratio. The functional substitution propensity score is obtained from the above. The formula for calculating the functional substitution propensity score is expressed as follows: in, This represents the functional replacement tendency value of the u-th initiated update block; This represents the ratio of commercial land area to residential land area in the current detailed control plan of the u-th block to be updated; This represents the daytime resident population of the u-th street that initiates the update; This represents the nighttime occupancy of the uth street where the update is initiated; This represents the land consolidation potential index for the u-th block to be updated.

[0037] All updated blocks will be activated sequentially, and the functional replacement tendency values ​​obtained from each block will be calculated. Arranged into a vector, each component of the vector corresponds to a starting update block, forming a functional replacement demand vector.

[0038] See Figure 4 In the figure, the horizontal axis represents the sequence number of the block to be updated, ranging from 1 to approximately 540, and the vertical axis represents the functional replacement tendency value corresponding to each block. The numerical range is from 0 to above 3.5. Different colored areas in the diagram correspond to different functional orientation categories: strong commercial orientation (dark red), weak commercial orientation (orange), functionally balanced (light yellow), weak residential orientation (light blue), and strong residential orientation (dark blue). The functional replacement orientation value is determined by the ratio of commercial land to residential land area in the current detailed control plan. The ratio of daytime to nighttime resident population difference and the product of the land plot integration potential index are used to determine the total population.

[0039] The chart shows that before the update of the street block sequence, the functional replacement tendency values ​​for approximately 270 blocks were generally in a high range, mainly concentrated in the strong commercial tendency and weak commercial tendency categories, with a peak value of about 3.5. Subsequently, the value gradually decreased, with a clear transition from the weak commercial tendency to the functional equilibrium range, where the value dropped to around 1. Blocks from approximately 270 to 360 were in the functional equilibrium range, with relatively small fluctuations and a generally flattened functional replacement tendency value. Blocks from 360 to 490 corresponded to a weak residential tendency, with the functional replacement tendency value continuously decreasing to around 0.7. At the end of the sequence, blocks from 490 to 540 were in the strong residential tendency range, with the functional replacement tendency value further decreasing to the lowest range, close to 0.5.

[0040] Example 4: In specific implementation, please refer to Figure 2 The process involves obtaining the existing building base outlines and existing building height distribution within each initiated update block. The existing building base outlines are extracted from the building census vector data, which stores the polygonal geometric information of the outer contour of the base of each existing building. These polygons are represented by ordered coordinate strings, and the existing building base outline of each initiated update block is a set of polygons representing the outer contours of the bases of all existing buildings within that block. The existing building height distribution includes the number of floors or the height value for each existing building. The number of floors is read from the building census attribute table, and the height value is calculated by multiplying the number of floors by the standard floor height, which is taken as 3.0 meters. If the building census attribute table directly provides a building height field, the building height field value is used directly.

[0041] Obtain the mode value of the current building height distribution. The mode value is the height value with the highest frequency among all current building height values ​​within the block where the update is initiated. Discretize the building height values ​​with equal width intervals of 0.5 meters, count the number of buildings in each height interval, select the height interval with the most buildings, and take the median value of that height interval as the mode value. If there are multiple height intervals with the same number of buildings, select the median value of the interval with the largest height value as the mode value.

[0042] The development intensity adjustment coefficient is calculated based on the functional replacement demand vector and the current control detailed planning indicators. The current control detailed planning indicators include the benchmark plot ratio and the maximum allowable plot ratio correction factor for each initiated urban renewal block. The benchmark plot ratio is the control benchmark value for the plot ratio of plots within the block under the current control detailed planning, and the maximum allowable plot ratio correction factor is the maximum multiple by which the plot ratio is allowed to increase to accommodate public facilities or policy-based housing allocation needs. The functional replacement tendency value corresponding to the initiated urban renewal block in the functional replacement demand vector is denoted as... Where the subscript d is the sequence number of the block in the update sequence. The maximum allowable floor area ratio correction factor is denoted as... , The value ranges from greater than 1.0 to a maximum of 1.5. Specific values ​​can be obtained from the current control plan development control table for the corresponding block. The formula for calculating the development intensity adjustment coefficient is: in, This represents the development intensity adjustment coefficient for the d-th block to be updated; This represents the functional replacement tendency value of the d-th initiated update block; This represents the maximum allowable floor area ratio correction factor in the current detailed control plan for the d-th block to be updated; This represents the normalized base floor area ratio of the d-th block that initiates the update. The calculation method is to divide the base floor area ratio of the d-th initiated renewal block by the maximum value of the base floor area ratios of all initiated renewal blocks, such that... The value of is between 0 and 1. When When it tends towards 1, Mainly affected The driving force, reflecting a strong demand for functional replacement, led to an increase in the floor area ratio; when When it approaches 0, It is mainly linked to the benchmark floor area ratio and tends to maintain the current control indicators.

[0043] Adjust the development intensity coefficient The updated building footprint area is obtained by multiplying the area of ​​the existing building footprint outline by the area of ​​the existing building footprint outline polygon. The area of ​​the existing building footprint outline is the projected area of ​​the polygonal shape of the existing building's base, calculated using a GIS area measurement tool by integrating the spherical area of ​​the polygon. For blocks containing multiple existing buildings, the area of ​​the existing building footprint outline is the sum of the areas of all existing building footprints. The development intensity adjustment factor is then applied. Multiplying the height by the modulo of the current building height distribution yields the updated building reference height.

[0044] Based on the updated building footprint area and updated building reference height, preliminary building blocks are generated for each block undergoing update, under the constraints of the block's internal topography and building setbacks. The block's internal topography constraints are extracted from the digital elevation model, while the building setback constraints are determined based on the building setback distances from the road red line and plot boundaries as specified in the current detailed control plan. The existing building footprint outline is scaled proportionally to the ratio of the updated building footprint area to the total area of ​​the existing building footprint, with the scaling center being the centroid of the existing building footprint outline. The scaled polygon serves as the base polygon of the preliminary building block. The preliminary building block is a right prism, with the base being the scaled polygon, the top being the polygon resulting from a vertical translation of the base to the updated building reference height, and the sides being extruded quadrilateral surfaces.

[0045] The spatial resilience coefficients at corresponding locations in the urban spatial resilience coefficient matrix are used as elevation constraints. The spatial resilience coefficient matrix, obtained from previous steps, contains the spatial resilience coefficients of each spatial grid. For the d-th block to be updated, the spatial resilience coefficients of all spatial grids within the block are obtained through spatial overlay. The median of these spatial resilience coefficients is taken as the representative spatial resilience coefficient of the block, denoted as . The representative spatial resilience coefficients of all initiated update blocks are subjected to max-min normalization to obtain normalized spatial resilience coefficients. The normalization formula is ,in, and These represent the minimum and maximum values ​​of the spatial resilience coefficient for all initiated update blocks, respectively. The normalized spatial resilience coefficient is then calculated. Mapped to height reduction factor The mapping method is This limits the height reduction factor to a range of 0.3 to 1.0. 0.3 is the preset lower limit for basic reduction, and 0.7 is the adjustable range. The basis for these settings is: collecting corresponding samples of spatial resilience coefficients and actual buildable heights from historical update projects, fitting the slope and intercept through linear regression, and determining the lower limit to 0.3 using rounding and conservative principles to ensure that the height of building blocks in areas with extremely low resilience is significantly compressed. The top elevation of the initial building block is then reduced and corrected; the height of the block after reduction and correction is equal to the updated building reference height multiplied by the height reduction factor. The bottom surface of the building block remains unchanged, while the top surface is regenerated vertically according to the corrected height of the building block, forming a three-dimensional spatial construction simulation outline. This three-dimensional spatial construction simulation outline is a three-dimensional geometric model of the preliminary building blocks of the block after height reduction correction, expressed as a three-dimensional solid.

[0046] Example 5: In practice, the planar location, diameter, and burial depth data of underground pipelines in each block of the updated block sequence are extracted from the comprehensive underground pipeline database. This database, stored in the urban municipal infrastructure management information system, contains spatial vector data and attribute fields for water supply pipelines, drainage pipelines, gas pipelines, heating pipelines, power pipelines, and communication pipelines. The planar location of underground pipelines is recorded as a polyline or smooth line segment in two dimensions. Each polyline consists of a sequentially arranged string of coordinate points, using the city's local coordinate system. Pipe diameter data is read from the attribute fields of the corresponding pipeline segment. For circular cross-section pipelines, the diameter is the inner diameter of the cross-section; for rectangular cross-section pipelines, the diameter is the equivalent inner diameter, which is equal to the diameter of the circle equivalent to the area of ​​the rectangular cross-section. Burial depth data is also read from the attribute fields of the corresponding pipeline segment; the burial depth is the vertical distance from the vertex of the pipeline's central axis to the ground surface.

[0047] For each underground pipeline, a spatial envelope model is generated. This model is a three-dimensional cylinder with the pipeline centerline as its axis and the pipe diameter as its diameter. Multiple segments of the pipeline centerline are vertically settled, with the settlement amount representing the burial depth, resulting in the three-dimensional centerline after burial depth. Using each node on the three-dimensional centerline as the center, a circular cross-section with a radius half the pipe diameter is generated in a plane perpendicular to the axis. All these circular cross-sections are swept along the three-dimensional centerline to form the three-dimensional cylinder. The top surface of the three-dimensional cylinder is a spatial surface formed by connecting the vertices of the shallowest circular cross-section, and its elevation corresponds to the pipe top burial depth. The pipe top burial depth is equal to the burial depth data minus half the pipe diameter.

[0048] The base extent of each building block in the 3D spatial construction simulation outline is overlaid with the horizontal projection extent of the underground pipeline space envelope model. The base extent of the building block is the vertical projection polygon of the building block's base polygon onto the horizontal plane. The horizontal projection extent of the underground pipeline space envelope model is the vertical projection area of ​​the 3D cylinder onto the horizontal plane, obtained by removing the elevation coordinates from the vertex coordinates of all triangular facets of the 3D cylinder to form a 2D convex hull. A 2D spatial intersection operation is performed on the base extent polygon and the horizontal projection extent polygon. If the intersection area is positive, the intersection area is recorded as a horizontal projection overlap area, represented by an intersecting polygon, and its associated building block number and underground pipeline number are recorded.

[0049] For each horizontally projected overlapping area, the foundation embedment depth of the building block is compared with the pipe top embedment depth value of the underground pipeline spatial envelope model. The foundation embedment depth is obtained from the foundation structural design parameters of the building block, which are preset based on the updated building reference height and foundation conditions of the building block. For multi-story building blocks with an updated building reference height not exceeding 24 meters, the foundation embedment depth is taken as 1.5 meters; for high-rise building blocks with an updated building reference height exceeding 24 meters, the foundation embedment depth is taken as one-tenth of the building reference height and not less than 3.0 meters. The pipe top embedment depth value is the elevation of the top surface of the three-dimensional cylinder at the location of the corresponding pipe segment in the underground pipeline spatial envelope model in the overlapping area, and is taken as the elevation value of the top surface point of the three-dimensional cylinder at the center point of the overlapping area.

[0050] If the foundation burial depth is greater than or equal to the pipe top burial depth, the overlapping area of ​​the horizontal projection is marked as failing the collision check; if the foundation burial depth is less than the pipe top burial depth, the overlapping area of ​​the horizontal projection is marked as passing the collision check. The marking process uses the following binary judgment: in, Indicates the first The building block and the first The constructability verification result of the underground pipeline in the horizontally overlapping area is marked with a value. A value of 1 indicates that the collision check failed. A value of 0 indicates that the collision check is successful. Indicates the first The foundation embedment depth of each building block is expressed in meters. Indicates the first The building block and the first The buried depth of the underground pipeline in the horizontally overlapping area is expressed in meters.

[0051] The constructability verification results are obtained by summarizing all the marked results. The constructability verification results are the combinations of all building blocks and underground pipelines. A set of values. If any one of the constructibility check results exists... If the value is 1, it is determined that there is a collision conflict in the corresponding block where the update is initiated.

[0052] Based on the constructability verification results, the sequence of districts to be initiated for urban renewal is adjusted, and an urban renewal planning scheme is output. Adjustment methods include: removing districts containing areas failing collision verification from the initiated urban renewal district sequence and downgrading these districts to the subsequent reserve urban renewal district list; or, for districts containing areas failing collision verification, keeping them within the initiated urban renewal district sequence, but rearranging the building blocks in their 3D spatial construction simulation outlines. The rearrangement involves shifting the planar position of the building blocks along a direction perpendicular to the pipeline direction until the building block base area and the horizontal projection area of ​​the underground pipeline spatial envelope model do not overlap, and the shift distance does not exceed half the corresponding side length of the building base outline. After adjustment, a final sequence of districts to be initiated for urban renewal and corresponding 3D spatial construction simulation outlines are formed. The initiated urban renewal district sequence, the functional replacement demand vector of each district, the development intensity adjustment coefficient, the 3D spatial construction simulation outlines, and the constructability verification results are integrated into an urban renewal planning scheme, and output in the form of Geographic Information System engineering documents and text reports.

[0053] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for urban renewal planning based on big data analysis, characterized in that, The method includes: Based on time-series data from the urban geographic information system, the trajectory of urban land use type changes and the sequence of infrastructure load fluctuations are extracted. Based on the land use type change trajectory and infrastructure load fluctuation sequence, an urban spatial resilience coefficient matrix is ​​constructed, and the set of blocks to be updated is located based on the urban spatial resilience coefficient matrix. Obtain the property rights subdivision map of each block in the block set to be updated, and generate a land consolidation potential index based on the property rights subdivision map and the current plot ratio distribution; Based on the land consolidation potential index, the set of blocks to be updated is hierarchically screened to determine the sequence of blocks to be updated. Collect the real-time population heat distribution and public transportation card swipe records corresponding to the initiated and updated street sequence, and construct a population flow time series tensor; Based on the population flow time series tensor and the land parcel integration potential index, generate the functional replacement demand vector for each initiated renewal block; Based on the aforementioned functional replacement demand vector and the current control detailed planning indicators, calculate the development intensity adjustment coefficient for each initiated renewal block; Based on the development intensity adjustment coefficient and the urban spatial resilience coefficient matrix, a three-dimensional spatial construction simulation outline is generated for each initiated renewal block. The three-dimensional spatial construction simulation outline is compared with the underground pipeline integrated database for collision detection to obtain the constructability verification results. Based on the constructability verification results, adjust the sequence of streets to be updated and output the urban renewal planning scheme.

2. The urban renewal planning method based on big data analysis according to claim 1, characterized in that, The step of extracting urban land use type change trajectories and infrastructure load fluctuation sequences based on time-series data from urban geographic information systems includes: The land use classification raster data of the city geographic information system at multiple historical time sections are acquired, the land use classification raster data of each time section are overlaid, the land use type change chain of the same geographic unit at different time sections is identified, and the land use type change trajectory is obtained. The system acquires infrastructure operation records from the city's geographic information system at multiple historical time points, extracts load values ​​of the same infrastructure node at each time point, and arranges the load values ​​in chronological order to obtain the infrastructure load fluctuation sequence.

3. The urban renewal planning method based on big data analysis according to claim 1, characterized in that, Based on the land use type change trajectory and infrastructure load fluctuation sequence, an urban spatial resilience coefficient matrix is ​​constructed, and the set of blocks to be updated is located based on the urban spatial resilience coefficient matrix, including: The urban area is divided into equal-sized spatial grids. The land use type change trajectory corresponding to each grid is encoded into a change vector, and the infrastructure load fluctuation sequence corresponding to each grid is encoded into a load vector. Calculate the first deviation between the transition vector of each grid and the preset stable land use pattern vector, and calculate the second deviation between the load vector of each grid and its historical mean vector. The first deviation and the second deviation are weighted and summed to obtain the spatial resilience coefficient of each grid. The spatial resilience coefficients of all grids are arranged to form the urban spatial resilience coefficient matrix. Extract the grids in the urban spatial resilience coefficient matrix whose spatial resilience coefficient is lower than a preset resilience threshold, and mark the blocks where the extracted grids are located as the set of blocks to be updated.

4. The urban renewal planning method based on big data analysis according to claim 1, characterized in that, The process of obtaining a detailed property rights map of each block in the set of blocks to be updated, and generating a land consolidation potential index based on the detailed property rights map and the current floor area ratio distribution, includes: Retrieve the cadastral database records of each block in the block set to be updated, parse the ownership boundary vector data of all property plots in each block, and generate the property plot subdivision map; Extract the boundary adjacency relationship and ownership information between adjacent property plots in the property plot subdivision map, and obtain the current plot ratio of each property plot; Based on the aforementioned boundary adjacency relationship, count the number of adjacent property plots with the same ownership entity within each block, and calculate the variance of the plot ratio distribution of all property plots within each block; The sum of the normalized quantities and the variance of the plot ratio distribution yields the land consolidation potential index for each block.

5. The urban renewal planning method based on big data analysis according to claim 1, characterized in that, The step of hierarchically filtering the set of blocks to be updated based on the land consolidation potential index to determine the sequence of blocks to be updated includes: The blocks in the block set to be updated are initially sorted from largest to smallest according to the land integration potential index; Based on the total number of blocks in the block set to be updated, a hierarchical interval step size is set, and each block is sequentially divided into multiple potential levels according to the results of the initial sorting. Calculate the average land consolidation potential index of all blocks in each potential tier, and mark all blocks in the potential tier with the highest average land consolidation potential index as the first batch of candidate blocks; The spatial distribution coordinates of the first batch of candidate blocks are overlaid with the existing key urban development axes for analysis. Candidate blocks located within the convergence area of ​​the key development axes are selected to form the block sequence to be updated.

6. The urban renewal planning method based on big data analysis according to claim 1, characterized in that, The process of collecting real-time population heatmaps and public transportation card swipe records corresponding to the initiated update block sequence and constructing a population flow time series tensor includes: The population heat map grid of each block in the initiated update block sequence is obtained from the mobile operator signaling data, and the grid values ​​are arranged in time sequence to generate a population heat map time series matrix. The system obtains the entry and exit card swipe records of bus stops and rail stations around the initiated and updated street sequence from the public transportation automatic fare collection system, and statistically analyzes the time-series vector of passenger flow for each station in chronological order. The population heat map time series matrix of all grids in the same block is concatenated with the passenger flow time series vector of each station in the block along the time dimension to form the population flow matrix of each block. The population flow matrices within all blocks are stacked along the block dimension to construct the population flow time series tensor.

7. The urban renewal planning method based on big data analysis according to claim 1, characterized in that, The step of generating functional replacement demand vectors for each initiated urban renewal block based on the population flow time series tensor and the land parcel integration potential index includes: The population flow time series tensor is decomposed into a periodic value in the time dimension to extract the population occupancy ratio of each block during the daytime and nighttime periods. Calculate the difference ratio between the daytime and nighttime resident population of each block, and multiply the difference ratio by the land consolidation potential index to obtain the commercial and residential function offset of each block. Obtain the area ratio of commercial land and residential land in the current planning of each block, and superimpose the commercial and residential function offset onto the area ratio to obtain the function replacement tendency value. The function replacement tendency values ​​of each initiated update block are arranged in block order to form the function replacement demand vector.

8. The urban renewal planning method based on big data analysis according to claim 1, characterized in that, The step of generating a three-dimensional spatial construction simulation outline for each initiated urban renewal block based on the development intensity adjustment coefficient and the urban spatial resilience coefficient matrix includes: Obtain the current building base outline and current building height distribution within each initiated update block; Multiply the development intensity adjustment coefficient by the area of ​​the existing building base outline to obtain the updated building base area; multiply the development intensity adjustment coefficient by the mode of the existing building height distribution to obtain the updated building reference height. Based on the updated building base area and the updated building reference height, a preliminary building block is generated for each block; Using the spatial resilience coefficient at the corresponding position in the urban spatial resilience coefficient matrix as an elevation constraint, the top surface elevation of the preliminary building block is reduced and corrected to generate the three-dimensional spatial construction simulation outline.

9. The urban renewal planning method based on big data analysis according to claim 1, characterized in that, The step of performing collision detection between the three-dimensional spatial construction simulation outline and the underground pipeline integrated database to obtain constructability verification results includes: The planar location, diameter, and burial depth data of underground pipelines in each block of the initiated update block sequence are extracted from the underground pipeline integrated database to generate an underground pipeline spatial envelope model. The base extent of each building block in the three-dimensional spatial construction simulation outline is overlaid with the horizontal projection extent of the underground pipeline space envelope model, and the horizontal projection overlap area is marked. For each horizontal projection overlap area, compare the foundation embedment depth of the building block with the pipe top embedment depth value of the underground pipeline spatial envelope model; If the foundation burial depth is greater than or equal to the pipe top burial depth, the overlapping area is marked as failing the collision check; otherwise, it is marked as passing the collision check. All marking results are then summarized to obtain the constructability check result.

10. A city renewal planning system based on big data analysis, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the urban renewal planning method based on big data analysis as described in any one of claims 1 to 9.

Images