Urban form iteration generation system and method based on high-frequency interaction space self-learning

Abstract

The invention discloses a city form iteration generation system and method based on high-frequency interaction space self-learning, and the system comprises a first data collection module which is used for obtaining a target city space form three-dimensional vector model in a current state; the second data acquisition module is used for constructing a case city spatial form three-dimensional vector model database; the graph network module is used for constructing associated network data of the target city spatial form three-dimensional vector model and an associated network database of the case city spatial form three-dimensional vector model database; a multi-target high-frequency iteration module; and a feasibility scheme extraction and display module. According to the method, multi-target-oriented generation iteration of the urban spatial form is realized, high-frequency interactive self-learning of urban spatial form scheme generation is realized, the rationality of the urban spatial form scheme is improved, and the problems of high manpower and material resource investment, human brain judgment, high randomness and small scale in the traditional urban design are avoided.

Description

Technical Field

[0001] The present invention belongs to the field of urban design application transformation, and in particular relates to an urban form iterative generation system and method based on high-frequency interactive space self-learning. Background Art

[0002] Urban form generation is one of the core tasks in urban design. Good urban form plays a vital role in urban planning and development. It is not only related to the quality of life of residents and determines the shaping of the city's image, but also affects the sustainable development of the city.

[0003] Traditional urban form design methods are usually based on historical experience, planning principles and local characteristics, and gradually improve the urban layout through a series of design steps. This results in low efficiency of urban form generation, limited number of iterations of urban form generation, and high labor and economic costs. Summary of the invention

[0004] Purpose of the invention: The first purpose of the present invention is to provide an iterative generation method of urban morphology based on high-frequency interactive space self-learning, which is efficient and can achieve high-frequency iteration under multi-objective guidance.

[0005] The second object of the present invention is to provide an urban morphology iterative generation system based on high-frequency interactive space self-learning.

[0006] Technical solution: The invention discloses an iterative generation method of urban form based on high-frequency interactive space self-learning, comprising the following steps:

[0007] S1: Obtain the three-dimensional vector model of the target city space form in the current state;

[0008] S2: Collect multiple case city spatial morphology 3D model data and build a case city spatial morphology 3D vector model database;

[0009] S3: Take the plots and buildings in the city as nodes, and the associations between plots, between plots and buildings belonging to the plots, and between buildings in the same plot as edges to construct a graph network, screen the required indicators in the urban space from the three dimensions of scale, shape and height, and convert the data extracted by quantifying the required indicators into node attributes; obtain the association network data of the target city spatial morphology 3D vector model and the association network database of the case city spatial morphology 3D vector model database;

[0010] S4: Based on the associative network database of the case urban spatial morphology three-dimensional vector model database, the required indicators are screened from the three dimensions of scale, morphology and height, and multiple goal orientations for iterative generation of urban morphology are constructed;

[0011] S5: The association network data of the target city spatial morphology three-dimensional vector model is input into the Neo4J platform. The Neo4J platform performs multiple iterations on the data under multiple goal-oriented reinforcement learning to obtain multiple goal-oriented urban spatial morphology solutions and construct a multi-goal-oriented urban spatial morphology association network solution set.

[0012] S6: Based on the multi-objective orientation, the urban spatial form association network scheme set is divided into multiple scheme groups, and the schemes in the three scheme groups are clustered from the required indicators contained in the three dimensions of scale, form, and height to obtain the feasible schemes corresponding to each group of schemes;

[0013] S7: Convert multiple feasible solutions into a three-dimensional vector space morphological model and output it for display.

[0014] Furthermore, the case city spatial morphology three-dimensional model data described in step S2 refers to the three-dimensional vector information of buildings including geographic coordinates, road contours or road red lines, plot boundaries, and plot plane functional division information.

[0015] Furthermore, the association relationship described in step S3 includes a distance relationship, an orientation relationship and a position relationship, and the attributes of the edges of the graph network include a distance attribute, an orientation attribute and a position attribute.

[0016] Furthermore, the distance attribute refers to the distance between the center points of two nodes, the orientation attribute refers to the orientation angle between the two nodes, and the position attribute refers to whether the two nodes are separated or connected.

[0017] Furthermore, the required indicators described in step S3 refer to: the scale dimension includes the plot area, total building area and floor area ratio; the morphological dimension includes the shape index, fractal dimension, compactness, enclosure and street-side building ratio; the height dimension includes the standard deviation and average height of the building height.

[0018] Furthermore, the multiple goal orientations described in step S3 include central development, cluster development and axis development.

[0019] Furthermore, central development, cluster development and axis development refer to:

[0020] Extracting the quantitatively extracted data of the required indicators of each case city space form three-dimensional vector model screened from three dimensions of scale, shape and height from the associated network database of the case city space form three-dimensional vector model database in step S3;

[0021] The quantitatively extracted data of all the required indicators are normalized to obtain the normalized data set x'. The normalization formula is as follows:

[0022]

[0023] In the formula, x ij It refers to the value of the i-th case city on the j-th required indicator;

[0024] Calculate the weight w corresponding to each required indicator j , the calculation formula is as follows:

[0025]

[0026] In the formula, y ij refers to the proportion of the i-th case city under the j-th required indicator, e j is the entropy value of the jth required indicator, w j is the weight of the jth required indicator;

[0027] The extracted data are quantified based on the normalized required indicators after weight calculation. The plots with the top 10% scale, height and shape in the extracted data are extracted after the required indicators are calculated. Adjacent plots are clustered to obtain clusters. The clusters with more than 15 plots in the clusters are defined as core clusters.

[0028] If a certain spatial morphology three-dimensional vector model has only one core cluster, it is judged as central development;

[0029] If a spatial morphology three-dimensional vector model has multiple non-adjacent core clusters, it is judged to be a cluster development;

[0030] If a spatial morphology three-dimensional vector model has multiple adjacent core area clusters, it is judged to be an axis type;

[0031] Calculate the graph network of the three-dimensional vector model of the spatial morphology of central development, cluster development and axis development respectively.

[0032] Furthermore, the method of obtaining multiple goal-oriented urban space form solutions in step S5 is as follows:

[0033] The Neo4J platform iterates it multiple times under a certain goal-oriented reinforcement learning, and each iteration generates a new spatial morphological association network. The spatial morphological association network generated by the nth iteration is A-NET n , A-NET n A-NET generated in the previous iteration n-1 Interactive comparison, and A-NET n The unreasonable plots in the A-NET n-1 The state in the process; set the iteration stop condition, after which multiple urban space morphology solutions oriented to the goal are obtained.

[0034] Furthermore, the unreasonable plots refer to plots whose changes in A-NETn exceed 20%.

[0035] Based on the same inventive concept, the present invention also discloses an urban form iterative generation system based on high-frequency interactive space self-learning, comprising:

[0036] The first data acquisition module is used to obtain the three-dimensional vector model of the target city space form in the current state;

[0037] The second data collection module is used to collect a plurality of case city space morphology three-dimensional model data and construct a case city space morphology three-dimensional vector model database;

[0038] A graph network module is used to construct a graph network of the target city spatial morphology three-dimensional vector model and each city spatial morphology vector model in the case city spatial morphology three-dimensional vector model database, and obtain the associated network data of the target city spatial morphology three-dimensional vector model and the associated network database of the case city spatial morphology three-dimensional vector model database; wherein the plots and buildings in the city are nodes of the graph network, and the associations between the plots, between the plots and the buildings belonging to the plots, and between the buildings in the same plot are edges of the graph network, and the required indicators in the urban space are screened from the three dimensions of scale, shape and height, and the data extracted by quantifying the required indicators are converted into attributes of the nodes;

[0039] The multi-objective high-frequency iteration module is used to iterate the associated network data of the target urban spatial morphology three-dimensional vector model multiple times under multiple objective-oriented reinforcement learning to obtain a multi-objective-oriented urban spatial morphology associated network solution set; wherein the multiple objective-oriented is based on the associated network database of the case urban spatial morphology three-dimensional vector model database, and is constructed by selecting the required indicators from the three dimensions of scale, morphology, and height;

[0040] The feasibility scheme extraction and display module is used to divide the multi-objective-oriented urban space form association network scheme set into multiple scheme groups, cluster the schemes in the three scheme groups according to the required indicators contained in the three dimensions of scale, form and height, obtain the feasibility scheme corresponding to each group of schemes, and convert multiple feasible schemes into three-dimensional vector space form models and output them for display.

[0041] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: the present invention, through indicator construction and reinforcement learning algorithm, constructs the three goal orientations of central development, cluster development and axis development for the generation of urban space form with a combination of three-dimensional indicators of scale, shape and height, thereby realizing the multi-goal-oriented generation iteration of urban space form; the present invention performs multiple iterations through goal-oriented reinforcement learning, thereby realizing high-frequency interactive self-learning for the generation of urban space form schemes, which is not only conducive to improving the rationality of urban space form schemes, but also avoids the problems of traditional urban design that require large manpower and material resources, involve human brain judgment, are highly arbitrary and small in scale. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of the method of the present invention;

[0043] Figure 2 A graph network of a three-dimensional vector model of the target city space form of the present invention;

[0044] Figure 3 It is a schematic diagram of the city center development of the present invention;

[0045] Figure 4 This is a schematic diagram of the urban cluster development of the present invention;

[0046] Figure 5 It is a schematic diagram of the city axis development of the present invention;

[0047] Figure 6 This is a translation diagram of the urban design guidelines for the plot of land for the feasible solution of the present invention.

[0048] Figure 7 Schematic diagram of the system of the present invention. DETAILED DESCRIPTION

[0049] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0050] Example 1

[0051] The urban form iterative generation method based on high-frequency interactive space self-learning described in the present invention is as follows: Figure 1 As shown, the following steps are included:

[0052] S1: Obtain the three-dimensional vector model of the target city's spatial form in the current state. Preferably, the three-dimensional vector model of the target city's spatial form is preprocessed, and the preprocessing includes unifying the dimensions of the three-dimensional vector model of the target city's spatial form, retaining the model data of the three-dimensional vector model of the target city's spatial form, cleaning the model data, clearing the remaining data in the model data, deleting duplicate data in the model data, and supplementing the missing data in the model data. The model data includes the height of the building in the target city, the area of ​​the building, the coordinate points of each vertex of the building, the width of the road, the coordinates of the center line of the road, and the coordinates of each vertex of the plot.

[0053] S2: Collect multiple case city spatial morphology 3D model data and build a case city spatial morphology 3D vector model database. The case city spatial morphology 3D model data refers to the building 3D vector information including geographic coordinates, road contours or road red lines, plot boundaries, and plot plane function division information.

[0054] S3: Take the plots and buildings in the city as nodes, and the relationships between plots, between plots and buildings belonging to the plots, and between buildings in the same plot as edges to construct a graph network, screen the required indicators in the urban space from the three dimensions of scale, shape and height, and convert the data extracted by quantifying the required indicators into node attributes; obtain the associated network data of the target city's spatial morphology three-dimensional vector model and the associated network database of the case city's spatial morphology three-dimensional vector model database.

[0055] A graph network is constructed with plots and buildings as nodes, and the relationships between plots, between plots and buildings belonging to the plots, and between buildings in the same plot as edges. The relationships include distance relationships, orientation relationships, and position relationships.

[0056] The indicators of all three-dimensional vector information in the target city space are quantitatively extracted, and then the required indicators are selected from the three dimensions of scale, shape and height, and the data extracted from the quantitative extraction of the required indicators are converted into the attributes of the nodes; the association relationship of the three-dimensional vector model of the target city space form is quantitatively extracted, and the data extracted from the quantitative association relationship is converted into the attributes of the edge. The nodes and their attributes, the edges and their attributes are saved in the graph network, and the association network data of the three-dimensional vector model of the target city space form is obtained, such as Figure 2 As shown. That is, buildings and plots are used as nodes of the graph network, and the data extracted from the quantitative indicators of the three dimensions of scale, shape and height are converted into node attributes. The relationship between plots and plots, plots and buildings belonging to the plots, and buildings belonging to the same plot are used as edges, and their attributes include distance attributes, orientation attributes, and position attributes. Preferably, the distance attribute refers to the distance between the center points of two nodes, the orientation attribute refers to the azimuth between two nodes, and the position attribute refers to whether the two nodes are separated or connected.

[0057] Taking the city as an independent subject, all the three-dimensional vector information of a single case city in the case city spatial morphology three-dimensional vector model database is quantitatively extracted, the required indicators are selected from the three dimensions of scale, shape and height, and the data of the quantitatively extracted required indicators are converted into the attributes of the node. The association relationship of the single case city spatial morphology three-dimensional vector model is quantitatively extracted, and the data of the quantitatively extracted association relationship is converted into the attributes of the edge. The nodes and their attributes, the edges and their attributes are saved in the graph network to obtain the association network data of the single case city spatial morphology three-dimensional vector model. That is, buildings and plots are taken as nodes of the graph network, and the data of the quantitatively extracted required indicators selected from the three dimensions of scale, shape and height are converted into the attributes of the nodes. The association relationship between plots and plots, plots and buildings, and buildings and buildings is taken as an edge, and its attributes include distance attributes, orientation attributes, and location attributes. The association network data of all single case city spatial morphology three-dimensional vector models constitute the association network database of the case city spatial morphology three-dimensional vector model database.

[0058] This embodiment realizes the digital transformation of three-dimensional model spatial information through spatial computing technology, and realizes the structured knowledge integration and storage of three-dimensional model spatial information through graph technology, thereby ensuring the integrity and standardization of data and improving the accuracy and efficiency of calculation.

[0059] In this embodiment, the specific description of the required indicators for screening in the three dimensions of scale, shape, and height, as well as the formula for quantitatively extracting the required indicators after screening are shown in Table 1 below.

[0060] Table 1

[0061]

[0062]

[0063] S4: Based on the associative network database of the three-dimensional vector model database of the case urban spatial morphology, the required indicators are screened from the three dimensions of scale, morphology and height, and multiple target orientations are constructed for the iterative generation of urban morphology.

[0064] The required indicators are screened in three dimensions: scale, form, and height. The entropy weight method is used to determine the weight of each required indicator. Based on the weight combination of each required indicator, the three goal orientations of central development, cluster development, and axis development for the generation of urban spatial form are constructed.

[0065] (1) Extracting the quantitative data of the required indicators of each case city space morphology three-dimensional vector model in the associated network database of the case city space morphology three-dimensional vector model database in step S3, which is screened from three dimensions: scale, shape and height.

[0066] (2) The quantitatively extracted data of all required indicators contained in the three dimensions of scale, shape, and height are weighted using the entropy weight method.

[0067] First, the quantitatively extracted data of all required indicators are normalized to obtain the normalized data set x'. The normalization formula is as follows:

[0068]

[0069] In the formula, x ij It refers to the value of the i-th case city on the j-th required indicator;

[0070] Calculate the weight w corresponding to each required indicator j , the calculation formula is as follows:

[0071]

[0072] In the formula, y ij refers to the proportion of the i-th case city under the j-th required indicator, e j is the entropy value of the jth required indicator, w j is the weight of the jth required indicator. The obtained weight is used to measure the proportion of the required indicators contained in the three dimensions of scale, shape and height, explore the morphological characteristics of the plot, and finally obtain three groups of data: scale group, shape group and height group.

[0073] The extracted data are quantified based on the normalized required indicators after weight calculation. The plots with the top 10% scale, height and shape in the extracted data are extracted after the required indicators are calculated. Adjacent plots are clustered to obtain clusters. The clusters with more than 15 plots in the clusters are defined as core clusters.

[0074] (3) If a spatial morphology three-dimensional vector model has only one core cluster, it is judged as central development, such as Figure 3 As shown; if a spatial morphology three-dimensional vector model has multiple non-adjacent core clusters, it is judged to be a cluster development, such as Figure 4 As shown; if a spatial morphology three-dimensional vector model has multiple adjacent core area clusters, it is judged to be an axis type, such as Figure 5 shown.

[0075] (4) Calculate the graph network composed of the quantitatively transformed data and the correlation relationships of the required indicators contained in the three dimensions of scale, shape and height of the three-dimensional vector model of the spatial form of central development, cluster development and axis development respectively.

[0076] S5: Input the association network data of the target city spatial form three-dimensional vector model into the Neo4J platform, and the Neo4J platform iterates it multiple times under multiple goal-oriented reinforcement learning to obtain multiple goal-oriented urban spatial form solutions and construct a multi-goal-oriented urban spatial form association network solution set, that is, to achieve high-frequency interactive spatial self-learning.

[0077] Based on the association network data of the target city spatial morphology three-dimensional vector model in step S3, multiple iterations are performed on the Neo4J platform through goal-oriented reinforcement learning. Each iteration generates a new spatial morphology association network. The spatial morphology association network generated in the nth iteration is A-NET n , A-NET n A-NET generated in the previous iteration n-1 For interactive comparison, A-NET n The unreasonable plots in the A-NET n-1 The state in the table is continuously iterated to form a new urban spatial form association network. The spatial form association network formed by each interactive iteration is traversed and compared with the cases in the association network database of the case city spatial form three-dimensional vector model database. When the difference between the required indicators of the three dimensions of scale, form, and height and one of the case cities is less than 10%, the iteration is stopped. Otherwise, the iteration continues until the iteration conditions are met.

[0078] Multiple rounds of high-frequency iterations of the urban spatial morphology association network were carried out through reinforcement learning under the three goal-oriented approaches of central development, cluster development and axis development, and urban spatial morphology schemes oriented towards central development, cluster development and axis development were obtained respectively. All the urban spatial morphology schemes obtained constitute a multi-goal-oriented urban spatial morphology association network scheme set.

[0079] The basis for identifying unreasonable plots is to construct a graph network that uses goal-oriented reinforcement learning to generate a new spatial morphology association network A-NETn, and define the plots that change by more than 20% compared with the association network data of the target city's spatial morphology three-dimensional vector model as unreasonable plots.

[0080] S6: Divide the urban spatial morphology association network solution set into multiple solution groups based on multi-objective orientation, cluster the solutions in the three solution groups based on the required indicators contained in the three dimensions of scale, form, and height, and obtain the feasible solutions corresponding to each group of solutions. Divide the multi-objective-oriented urban spatial morphology association network solution set of the target city into three solution groups based on central development, cluster development, and axis development, cluster the solutions in the three solution groups based on the required indicators contained in the three dimensions of scale, form, and height, and the solutions obtained after clustering are the feasible solutions corresponding to each group of solutions. Preferably, the clustering method is K-means clustering.

[0081] S7: Convert the feasibility plans corresponding to central development, cluster development and axis development into three-dimensional vector space morphological models and output them for display.

[0082] Preferably, the feasible solutions are displayed in a sandbox and human-computer interaction is realized.

[0083] Establish a holographic projection sand table to interactively display the high-frequency iterative generation plan of urban space form under multi-objective guidance. The required equipment includes a three-dimensional physical sand table, a digital holographic projector and a gesture recognizer.

[0084] The interactive display process specifically includes using a projector with a resolution of 4K or above, a depth sensor, a motion sensor, a touch sensor, and a sand table to build a three-dimensional digital sand table, using a handheld controller to display the feasible solution in step S7 in a virtual scene, and using VR glasses and virtual reality gloves to achieve scene interaction with the model. Among them, the handheld controller is used to achieve an in-depth experience of the model; the virtual reality gloves are used to achieve the selection, scaling, and modification of the model; the VR glasses are used to achieve an interactive display of the model, and the model can be operated together with the virtual reality gloves.

[0085] Taking a certain area of ​​Chuzhou City as the target city, the three-dimensional model data of urban space morphology in Hangzhou, Nanjing, Wuxi, Suzhou, and Shanghai were collected to construct a three-dimensional vector model database of case city space morphology, and the urban morphology iterative generation method based on high-frequency interactive spatial self-learning of the present invention was implemented. The final feasible solution and the three-dimensional vector model of the target city space morphology in the current state were displayed in the translation diagram of the land urban design guidelines of the feasible solution, as shown in the figure. Figure 6 shown.

[0086] Example 2

[0087] The urban form iterative generation system based on high-frequency interactive space self-learning described in the present invention is as follows: Figure 7As shown, it includes a first data acquisition module, a second data acquisition module, a graph network module, a multi-objective high-frequency iteration module and a feasibility solution extraction and display module.

[0088] The first data acquisition module is used to obtain the three-dimensional vector model of the target city's spatial form in the current state. Preferably, the three-dimensional vector model of the target city's spatial form is preprocessed, and the preprocessing includes unifying the dimension of the three-dimensional vector model of the target city's spatial form, retaining the model data of the three-dimensional vector model of the target city's spatial form, cleaning the model data, clearing the remaining data in the model data, deleting the duplicate data in the model data, and supplementing the missing data in the model data. The model data includes the height of the building in the target city, the area of ​​the building, the coordinate points of each vertex of the building, the width of the road, the coordinates of the center line of the road, and the coordinates of each vertex of the plot.

[0089] The second data collection module is used to collect multiple case city spatial morphology 3D model data and build a case city spatial morphology 3D vector model database. The case city spatial morphology 3D model data refers to the building 3D vector information including geographic coordinates, road contour lines or road red lines, plot boundaries, and plot plane function division information.

[0090] The graph network module is used to construct the graph network of the target city spatial morphology three-dimensional vector model and each city spatial morphology vector model in the case city spatial morphology three-dimensional vector model database, and obtain the associated network data of the target city spatial morphology three-dimensional vector model and the associated network database of the case city spatial morphology three-dimensional vector model database; wherein the plots and buildings in the city are nodes of the graph network, and the association relationships between the plots, between the plots and the buildings belonging to the plots, and between the buildings in the same plot are the edges of the graph network, and the required indicators in the urban space are screened from the three dimensions of scale, shape and height, and the data extracted by quantitative extraction of the required indicators are converted into the attributes of the nodes. The required indicators refer to: the plot area, total building area and volume ratio included in the scale dimension; the morphological dimension includes shape index, fractal dimension, compactness, enclosure and street building ratio; the height dimension includes the standard deviation and average height of the building height. And the specific description of the required indicators and the formula for quantitative extraction of the required indicators are consistent with Table 1 in Example 1.

[0091] The indicators of all three-dimensional vector information in the target city space are quantitatively extracted, and then the required indicators are selected from the three dimensions of scale, shape and height, and the data extracted by the quantitative extraction of the required indicators are converted into the attributes of the nodes; the association relationship of the three-dimensional vector model of the target city space form is quantitatively extracted, and the data extracted by the quantitative extraction of the association relationship is converted into the attributes of the edge. The nodes and their attributes, the edges and their attributes are saved in the graph network to obtain the association network data of the three-dimensional vector model of the target city space form. That is, the buildings and plots are used as nodes of the graph network, and the data extracted by the quantitative extraction of the required indicators selected from the three dimensions of scale, shape and height are converted into the attributes of the nodes. The association relationship between the plots and the plots, the plots and the buildings, and the buildings and the buildings is used as the edge, and its attributes include distance attribute, orientation attribute, and position attribute. Preferably, the distance attribute refers to the distance between the center points of the two nodes, the orientation attribute refers to the azimuth between the two nodes, and the position attribute refers to whether the two nodes are separated or connected.

[0092] Taking the city as an independent subject, all the three-dimensional vector information of a single case city in the case city spatial morphology three-dimensional vector model database is quantitatively extracted, the required indicators are selected from the three dimensions of scale, shape and height, and the data of the quantitatively extracted required indicators are converted into the attributes of the node. The association relationship of the single case city spatial morphology three-dimensional vector model is quantitatively extracted, and the data of the quantitatively extracted association relationship is converted into the attributes of the edge. The nodes and their attributes, the edges and their attributes are saved in the graph network to obtain the association network data of the single case city spatial morphology three-dimensional vector model. That is, buildings and plots are taken as nodes of the graph network, and the data of the quantitatively extracted required indicators selected from the three dimensions of scale, shape and height are converted into the attributes of the nodes. The association relationship between plots and plots, plots and buildings, and buildings and buildings is taken as an edge, and its attributes include distance attributes, orientation attributes, and location attributes. The association network data of all single case city spatial morphology three-dimensional vector models constitute the association network database of the case city spatial morphology three-dimensional vector model database.

[0093] The multi-objective high-frequency iteration module is used to iterate the associated network data of the target urban spatial form three-dimensional vector model under multiple objective-oriented reinforcement learning to obtain a multi-objective-oriented urban spatial form associated network solution set; wherein the multiple objective-oriented is based on the associated network database of the case urban spatial form three-dimensional vector model database, and is constructed by selecting the required indicators from the three dimensions of scale, form, and height. That is, high-frequency interactive spatial self-learning is realized.

[0094] The required indicators are screened in three dimensions: scale, form, and height. The entropy weight method is used to determine the weight of each required indicator. Based on the weight combination of each required indicator, the three goal orientations of central development, cluster development, and axis development for the generation of urban spatial form are constructed.

[0095] (1) Extract the quantitative data of the required indicators of the three-dimensional vector model of the case city space form in the associated network database of the case city space form three-dimensional vector model in the graph network module, which are screened from the three dimensions of scale, shape and height.

[0096] (2) The quantitatively extracted data of all required indicators contained in the three dimensions of scale, shape, and height are weighted using the entropy weight method.

[0097] First, the quantitatively extracted data of all required indicators are normalized to obtain the normalized data set x'. The normalization formula is as follows:

[0098]

[0099] In the formula, x ij It refers to the value of the i-th case city on the j-th required indicator;

[0100] Calculate the weight w corresponding to each required indicator j , the calculation formula is as follows:

[0101]

[0102] In the formula, y ij refers to the proportion of the i-th case city under the j-th required indicator, e j is the entropy value of the jth required indicator, w j is the weight of the jth required indicator. The obtained weight is used to measure the proportion of the required indicators contained in the three dimensions of scale, shape and height, explore the morphological characteristics of the plot, and finally obtain three groups of data: scale group, shape group and height group.

[0103] The extracted data are quantified based on the normalized required indicators after weight calculation. The plots with the top 10% scale, height and shape in the extracted data are extracted after the required indicators are calculated. Adjacent plots are clustered to obtain clusters. The clusters with more than 15 plots in the clusters are defined as core clusters.

[0104] (3) If a spatial morphology three-dimensional vector model has only one core cluster, it is judged as central development; if a spatial morphology three-dimensional vector model has multiple non-adjacent core clusters, it is judged as cluster development; if a spatial morphology three-dimensional vector model has multiple adjacent core area clusters, it is judged as axis development.

[0105] (4) Calculate the graph network composed of the quantitatively transformed data and the correlation relationships of the required indicators contained in the three dimensions of scale, shape and height of the three-dimensional vector model of the spatial form of central development, cluster development and axis development respectively.

[0106] The association network data of the target city's spatial morphology three-dimensional vector model is input into the Neo4J platform. The Neo4J platform iterates it multiple times under multiple goal-oriented reinforcement learning to obtain multiple goal-oriented urban spatial morphology solutions and construct a multi-goal-oriented urban spatial morphology association network solution set.

[0107] Based on the association network data of the three-dimensional vector model of the target city spatial form in the graph network module, multiple iterations are performed on the Neo4J platform through goal-oriented reinforcement learning. Each iteration generates a new spatial form association network. The spatial form association network generated in the nth iteration is A-NET n , A-NET n A-NET generated in the previous iteration n-1 For interactive comparison, A-NET n The unreasonable plots in the A-NET n-1 The state in the table is continuously iterated to form a new urban spatial form association network. The spatial form association network formed by each interactive iteration is traversed and compared with the cases in the association network database of the case city spatial form three-dimensional vector model database. When the difference between the required indicators of the three dimensions of scale, form, and height and one of the case cities is less than 10%, the iteration is stopped. Otherwise, the iteration continues until the iteration conditions are met.

[0108] Multiple rounds of high-frequency iterations of the urban spatial morphology association network were carried out through reinforcement learning under the three goal-oriented approaches of central development, cluster development and axis development, and urban spatial morphology schemes oriented towards central development, cluster development and axis development were obtained respectively. All the urban spatial morphology schemes obtained constitute a multi-goal-oriented urban spatial morphology association network scheme set.

[0109] The basis for identifying unreasonable plots is to construct a graph network that uses goal-oriented reinforcement learning to generate a new spatial morphology association network A-NETn, and define the plots that change by more than 20% compared with the association network data of the target city's spatial morphology three-dimensional vector model as unreasonable plots.

[0110] The feasibility scheme extraction and display module is used to divide the multi-objective-oriented urban space form association network scheme set into multiple scheme groups, cluster the schemes in the three scheme groups according to the required indicators contained in the three dimensions of scale, form and height, obtain the feasibility scheme corresponding to each group of schemes, and convert multiple feasible schemes into three-dimensional vector space form models and output them for display.

[0111] According to the multi-objective orientation, the urban spatial form association network solution set is divided into multiple solution groups, and the solutions in the three solution groups are clustered from the required indicators contained in the three dimensions of scale, form, and height to obtain the feasible solutions corresponding to each group of solutions. The multi-objective-oriented urban spatial form association network solution set of the target city is divided into three solution groups according to central development, cluster development, and axis development. The solutions in the three solution groups are clustered from the required indicators contained in the three dimensions of scale, form, and height. The solutions obtained after clustering are the feasible solutions corresponding to each group of solutions. Preferably, the clustering method is K-means clustering.

[0112] The feasibility plans corresponding to central development, cluster development and axis development are converted into three-dimensional vector space morphological models and output and displayed.

[0113] Preferably, the feasible solutions are displayed in a sandbox and human-computer interaction is realized.

[0114] Establish a holographic projection sand table to interactively display the high-frequency iterative generation plan of urban space form under multi-objective guidance. The required equipment includes a three-dimensional physical sand table, a digital holographic projector and a gesture recognizer.

[0115] The interactive display process specifically includes using a projector with a resolution of 4K or above, a depth sensor, a motion sensor, a touch sensor, and a sand table to build a three-dimensional digital sand table, using a handheld controller to display the feasible solution in step S7 in a virtual scene, and using VR glasses and virtual reality gloves to achieve scene interaction with the model. Among them, the handheld controller is used to achieve an in-depth experience of the model; the virtual reality gloves are used to achieve the selection, scaling, and modification of the model; the VR glasses are used to achieve an interactive display of the model, and the model can be operated together with the virtual reality gloves.

Claims

1. A method for iterative generation of urban morphology based on high-frequency interactive space self-learning, characterized by: The following steps are involved: S1: Obtain the three-dimensional vector model of the target city space form in the current state; S2: Collect multiple case city spatial morphology 3D model data and build a case city spatial morphology 3D vector model database; S3: Take the plots and buildings in the city as nodes, and the relationships between plots, between plots and buildings belonging to the plots, and between buildings in the same plot as edges to build a graph network. Select the required indicators in the urban space from the three dimensions of scale, shape and height, and convert the data extracted by quantifying the required indicators into node attributes. Obtaining the associated network data of the target city's spatial morphology three-dimensional vector model and the associated network database of the case city's spatial morphology three-dimensional vector model database; S4: Based on the associative network database of the case urban spatial morphology three-dimensional vector model database, the required indicators are screened from the three dimensions of scale, morphology and height, and multiple goal orientations for iterative generation of urban morphology are constructed; S5: The association network data of the target city spatial morphology three-dimensional vector model is input into the Neo4J platform. The Neo4J platform performs multiple iterations on the data under multiple goal-oriented reinforcement learning to obtain multiple goal-oriented urban spatial morphology solutions and construct a multi-goal-oriented urban spatial morphology association network solution set. S6: Based on the multi-objective orientation, the urban spatial form association network scheme set is divided into multiple scheme groups, and the schemes in the three scheme groups are clustered from the required indicators contained in the three dimensions of scale, form, and height to obtain the feasible schemes corresponding to each group of schemes; S7: Convert multiple feasible solutions into a three-dimensional vector space morphological model and output it for display.

2. The urban form iterative generation method based on high-frequency interactive space self-learning according to claim 1 is characterized by: The case city spatial morphology three-dimensional model data described in step S2 refers to the building three-dimensional vector information including geographic coordinates, road contour lines or road red lines, plot boundaries, and plot plane functional division information.

3. The urban form iterative generation method based on high-frequency interactive space self-learning according to claim 1 is characterized by: The association relationship described in step S3 includes a distance relationship, an orientation relationship and a position relationship, and the attributes of the edges of the graph network include a distance attribute, an orientation attribute and a position attribute.

4. The urban form iterative generation method based on high-frequency interactive space self-learning according to claim 1 is characterized by: The distance attribute refers to the distance between the center points of two nodes, the orientation attribute refers to the orientation angle between the two nodes, and the position attribute refers to whether the two nodes are separated or connected.

5. The urban form iterative generation method based on high-frequency interactive space self-learning according to claim 1 is characterized by: The required indicators described in step S3 are: the scale dimension includes the plot area, total building area and floor area ratio; the morphological dimension includes the shape index, fractal dimension, compactness, enclosure and street-side building ratio; the height dimension includes the building height standard deviation and average height.

6. The urban form iterative generation method based on high-frequency interactive space self-learning according to claim 1 is characterized by: The multiple goal orientations described in step S3 include central development, cluster development and axis development.

7. The urban form iterative generation method based on high-frequency interactive space self-learning according to claim 6 is characterized by: Central development, cluster development and axis development refer to: Extracting the quantitatively extracted data of the required indicators of each case city space morphology three-dimensional vector model screened from three dimensions of scale, shape and height from the associated network database of the case city space morphology three-dimensional vector model database in step S3; The quantitatively extracted data of all the required indicators are normalized to obtain the normalized data set x'. The normalization formula is as follows: In the formula, x ij It refers to the value of the i-th case city on the j-th required indicator; Calculate the weight w corresponding to each required indicator j , the calculation formula is as follows: In the formula, y ij refers to the proportion of the i-th case city under the j-th required indicator, e j is the entropy value of the jth required indicator, w j is the weight of the jth required indicator; The extracted data are quantified based on the normalized required indicators after weight calculation. The plots with the top 10% scale, height and shape in the extracted data are extracted after the required indicators are calculated. Adjacent plots are clustered to obtain clusters. The clusters with more than 15 plots in the clusters are defined as core clusters. If a certain spatial morphology three-dimensional vector model has only one core cluster, it is judged as central development; If a spatial morphology three-dimensional vector model has multiple non-adjacent core clusters, it is judged to be a cluster development; If a spatial morphology three-dimensional vector model has multiple adjacent core area clusters, it is judged to be an axis type; Calculate the graph network of the three-dimensional vector model of the spatial morphology of central development, cluster development and axis development respectively.

8. The urban form iterative generation method based on high-frequency interactive space self-learning according to claim 1 is characterized by: The method of obtaining multiple goal-oriented urban space form solutions in step S5 is as follows: The Neo4J platform iterates it multiple times under a certain goal-oriented reinforcement learning, and each iteration generates a new spatial morphological association network. The spatial morphological association network generated by the nth iteration is A-NET n , A-NET n A-NET generated in the previous iteration n-1 Interactive comparison, and A-NET n The unreasonable plots in the A-NET n-1 Status in The iteration stopping condition is set, and after the iteration stops, multiple urban space morphology solutions oriented to the goal are obtained.

9. The urban form iterative generation method based on high-frequency interactive space self-learning according to claim 8 is characterized by: The unreasonable plots refer to the plots whose changes in A-NETn exceed 20%.

10. An iterative generation system of urban form based on high-frequency interactive space self-learning, characterized by: include, The first data acquisition module is used to obtain the three-dimensional vector model of the target city space form in the current state; The second data collection module is used to collect a plurality of case city space morphology three-dimensional model data and construct a case city space morphology three-dimensional vector model database; A graph network module is used to construct a graph network of the target city spatial morphology three-dimensional vector model and each city spatial morphology vector model in the case city spatial morphology three-dimensional vector model database, and obtain the associated network data of the target city spatial morphology three-dimensional vector model and the associated network database of the case city spatial morphology three-dimensional vector model database; wherein the plots and buildings in the city are nodes of the graph network, and the associations between the plots, between the plots and the buildings belonging to the plots, and between the buildings in the same plot are edges of the graph network, and the required indicators in the urban space are screened from the three dimensions of scale, shape and height, and the data extracted by quantifying the required indicators are converted into attributes of the nodes; The multi-objective high-frequency iteration module is used to iterate the associated network data of the target urban spatial morphology three-dimensional vector model multiple times under multiple objective-oriented reinforcement learning to obtain a multi-objective-oriented urban spatial morphology associated network solution set; wherein the multiple objective-oriented is based on the associated network database of the case urban spatial morphology three-dimensional vector model database, and is constructed by selecting the required indicators from the three dimensions of scale, morphology, and height; The feasibility scheme extraction and display module is used to divide the multi-objective-oriented urban space form association network scheme set into multiple scheme groups, cluster the schemes in the three scheme groups according to the required indicators contained in the three dimensions of scale, form and height, obtain the feasibility scheme corresponding to each group of schemes, and convert multiple feasible schemes into three-dimensional vector space form models and output them for display.

Images