Building group automatic layout method and system based on function-form interaction

Abstract

The invention discloses a building group automatic layout method and system based on function-form interaction, and the method comprises the following steps: collecting design boundary data, road network data, ecological environment data and existing building data of a target block, and constructing a three-dimensional space digital sand table; the method comprises the following steps: collecting building floor area, surrounding road attributes, building floor number and land use function data of a plurality of case blocks, and constructing a land use function and spatial form gray level co-occurrence matrix; performing factor analysis to obtain function-form factor load matrix file data, and generating a function-form correlation model based on the knowledge graph; performing function-form dynamic interaction building group generation by applying a depth deterministic strategy gradient algorithm, and generating a building group layout scheme; upper planning is input, and a screened 3D block building group model object set meeting the requirements of the upper planning is output; and superposing the screened 3D block building group model on a digital sand table, and determining and outputting a final scheme.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban planning, and in particular relates to an automatic layout method and system for building complexes based on function-form interaction. Background Art

[0002] With the continuous advancement of new urbanization construction, urban development has gradually entered a stage of high-quality development that focuses on spatial quality from the stage of pursuing incremental expansion. Urban space is the result of the comprehensive effect of two-dimensional functional layout and three-dimensional spatial form. Function and form, as the core components of the block space system, influence each other in space, couple and nest, and jointly create a good spatial order and living environment. In this context, the contradiction between the high-intensity architectural form and low-density functional vitality in urban blocks has become increasingly prominent. The simple pursuit of the "height" and "breadth" of form is likely to ignore the actual needs of functional development, resulting in unreasonable layout of production and living space, low operating efficiency and lack of spatial vitality, which in turn leads to a series of dilemmas such as spatial disorder and imbalance in the relationship between people and land, hindering the sustainable and healthy development of cities.

[0003] At present, the common research methods for building complex model layout mainly include automatic layout methods based on direct nesting of case libraries. In terms of research objects, they are limited to the design parameters of existing cases; in terms of research accuracy, due to cost constraints, the number of building complex case samples is limited and the recognition accuracy is low, making it difficult to cover various functional scenarios and design positioning; in terms of research conclusions, the direct nesting method is not enough to solve complex design problems limited by multiple design goals and constraints, which may lead to insufficient fit between the scheme and the target block.

[0004] Therefore, it is urgent to solve the above problems. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide an automatic layout method for building complexes based on function-form interaction, which can achieve highly coupled coordination between function and form, and an automatic layout method for block building complexes with more diverse arrangement combinations and higher design efficiency.

[0006] Technical solution: To achieve the above purpose, the present invention discloses an automatic layout method of a building complex based on function-form interaction, comprising the following steps:

[0007] (1) Collect the design boundary data, road network data, ecological environment data and existing building data of the target block, perform data processing and spatial superposition, and construct a three-dimensional digital sand table;

[0008] (2) Collect data on building area, surrounding road attributes, number of building floors and land use functions of several case blocks, process them to obtain all land use functions and spatial morphological index types of the building complexes in the case blocks, form a function-morphological case library, uniformly display all land use functions and spatial morphological index types in grayscale classification, and construct a grayscale symbiosis matrix of land use functions and spatial morphology;

[0009] (3) Based on the arrangement rules between the pixels of the gray-level co-occurrence matrix, factor analysis is performed to obtain the function-morphology factor loading matrix file data, and the factor loading matrix file data is imported into the graph database to complete the knowledge storage from the text file to the graph database. The relationship between variables and factors is provided with a graphical visualization function, and a function-morphology association model based on the knowledge graph is generated;

[0010] (4) Apply the deep deterministic policy gradient algorithm to generate building complexes with dynamic functional-morphological interaction, import the functional-morphological factor load matrix in step (3), and abstract it into the concepts of state, action, and reward; correspond the policy model output of the deep deterministic policy gradient algorithm to the control signal of the corresponding Grasshopper battery pack, adjust the functional-morphological parameters based on the block building complex model, receive instructions from the deep deterministic policy gradient algorithm, and update the state of the physical building complex model, batch output the functional-morphological parameters, and generate the building complex layout plans for business blocks, commercial blocks, residential blocks, public service blocks, and mixed blocks, that is, output the Rhinoceros model data of the 3D block building complex;

[0011] (5) Input the threshold range of building height, building density, and development intensity index for the target block form in the superior planning, connect the Rhinoceros model data of the 3D block building complex to the corresponding input port of the Query Model Objects screening component, and adjust the screening conditions in the component's property panel. The screening conditions are the threshold range of building height, building density, and development intensity index for the target block form in the superior planning. Run the definition to process the data and output the 3D block building complex model object set that meets the requirements of the superior planning after screening;

[0012] (6) Based on the digital sand table of the target block constructed in step (1), the selected 3D block building complex model is superimposed on the digital sand table; a voice recognition device and a mixed reality wearable 3D motion capture device are installed on the 3D holographic digital sand table to simulate the scheme and display indicators, and a human-computer interaction instruction library is constructed. The final scheme is determined and outputted through the mixed reality head-mounted device in the actual scene.

[0013] Optionally, in step (1), based on the GIS geographic information platform, the design boundary data, road network data, ecological environment data and existing building data of the target block are collected, and data processing and spatial superposition are performed in the ArcGis platform to construct a three-dimensional digital sand table; the ecological environment data includes mountain data, water body data, ecologically sensitive area data and open space data, and the existing building data includes latitude and longitude coordinates, building floor area and building number of floors.

[0014] Optionally, in step (2), the building area, number of building floors, and surrounding road attributes are retrieved from the target block range on the Open Street Map platform and then converted into a new format; the land use function data is calculated from the POI data using the TF-IDF algorithm, and the calculation formula is:

[0015]

[0016] Among them, v is the number of the block unit, w is a certain POI type, and k represents the t containing the POI to be retrieved. v The number of case blocks, S v,w represents the area of ​​type w land in the v block unit, n v,w is the frequency of type w business in the vth block unit, D represents the total number of all POIs, |j:t v ∈d w | indicates the number of v contained; there is a corresponding relationship between POI types and land use functions.

[0017] Optionally, the types of spatial morphological indicators in step (2) include quantitative indicators of scale, symbolism and diversity; among which scale indicators include total area, total base area, building density, floor area ratio and average height; symbolism indicators include maximum height and maximum base area; diversity indicators include height staggeredness and base area staggeredness; land use functions include production service functions, living service functions and public welfare service functions.

[0018] Optionally, the specific steps for processing in step (2) to obtain all land use functions and spatial morphological indicator types of the case block building complex are as follows: based on the spatial position coordinates in the ArcGIS platform, the building complex spatial morphological data obtained from the OSM platform and the case block unit data obtained by road network buffering are spatially linked to obtain the base outlines and building floor number information of all buildings in the case block, that is, all buildings in the same neighborhood share a neighborhood ID; and then, according to the calculation formulas for various indicators, the scale indicator, landmark indicator and diversity indicator of the spatial morphology of each block are calculated.

[0019] Optionally, in step (2), the gray-level symbiosis matrix of land use function and spatial morphology is constructed by using the imread function in MATLAB to input the spatial morphology data and land use function data of the case block building complex reclassified by the natural break point method; the gray-level symbiosis matrix is ​​calculated by the graycomatrix function, and the calculation formula of the gray-level symbiosis matrix correlation C is:

[0020]

[0021] Among them, N is the number of gray levels segmented by the natural break point method, P(p,q) is the value of each data of each case block building complex in the gray level co-occurrence matrix, p and q are the land use function of the case block building complex and the gray value of each data of each case block building complex, respectively; the gray level co-occurrence matrix is ​​used to measure the correlation between the pixel and its adjacent pixels in the image, reflecting the regularity and consistency of its texture; based on the frequency of occurrence of pixel pairs with specific gray value combinations in specific direction and distance data, the correlation characteristics between various spatial morphological data and land use functions are quantitatively calculated.

[0022] Optionally, in step (3), a factor loading matrix is ​​constructed based on the spatial morphological factors and land use function type factors of the case block, and the spatial morphological data and land use function data are exported as factor data and variable data respectively; a FactorAnalysis object is created through the scikit-learn library of Python, and the number of factors to be extracted is specified as 9 and the number of original variables is specified as 1; the spatial morphological factor data of the case block is imported as a NumPy array as the data set to be analyzed, the factor data is fitted using the fit_transform method, and the factor loading matrix of the three major functional business types of production service function, life service function, and public welfare service function is obtained using the components_ attribute; wherein the size of the load represents the correlation strength between the building functional business type variable and each factor, and the symbol represents the positive or negative relationship between them; the correlation relationship between the factors in the step (3) is based on the correlation structure of the factor loading matrix, and the corresponding factor nodes, variable nodes and relationship networks are created through Neo4jDesktop; the Neo4j Browser graphic visualization tool is used to design a visual representation of the factor loading matrix, run a Cypher query to create a graphic view, and adjust the visualization settings to render and export the graphic data set.

[0023] Optionally, in the deep deterministic policy gradient algorithm in step (4), the state is the current spatial morphological characteristics and land use function type of the block building complex, the action is the next design decision made according to the current state, which is achieved by modifying the morphological attributes and functional attributes of the block building; the reward is a method used to evaluate the degree of function-morphological matching of the block building through a reward function; based on the factor load values ​​of the corresponding functional business type and the arrangement rules between the pixels of the gray-level co-occurrence matrix, a corresponding reward function is set to evaluate the pros and cons of the action.

[0024] Optionally, in the function-morphology parameter adjustment loop in step (4), first input the multi-agent and environment chassis to obtain the initial state S0; at the tth step, randomly assign the spatial agent actions At and Bt to obtain the state St; calculate the benefit of St according to the reward function, and output the evaluation value Qt of a series of action strategies in the current round; store the current sequence (At, Bt, St, Qt) in the experience pool D, combine the experience replay and fixed Q network mechanism to reduce process fitting, and optimize the training to obtain a series of action strategies with the largest Q value.

[0025] Based on the same inventive concept, the present invention discloses an automatic layout system for building complexes based on function-form interaction, comprising:

[0026] The digital sand table construction module is used to collect the design boundary data, road network data, ecological environment data and existing building data of the target block, perform data processing and spatial superposition, and construct a three-dimensional digital sand table;

[0027] The gray-level co-occurrence matrix construction module is used to collect the building area, surrounding road attributes, number of building floors and land use function data of several case blocks, and process them to obtain all the land use functions and spatial morphological index types of the building complex in the case blocks, forming a function-morphological case library, and uniformly displaying all the land use functions and spatial morphological index types in gray-level classification to construct the land use function and spatial morphological gray-level co-occurrence matrix;

[0028] The function-morphology association model generation module is used to perform factor analysis based on the arrangement rules between the pixels of the gray-level co-occurrence matrix to obtain the function-morphology factor load matrix file data, import the factor load matrix file data into the graph database, complete the knowledge storage from the text file to the graph database, provide a graphical visualization function for the association between variables and factors, and generate a function-morphology association model based on the knowledge graph;

[0029] The building complex layout scheme generation module is used to apply the deep deterministic policy gradient algorithm to generate building complexes with dynamic functional-morphological interaction, import the functional-morphological factor load matrix, and abstract it into the concepts of state, action, and reward; the policy model output of the deep deterministic policy gradient algorithm corresponds to the control signal of the corresponding Grasshopper battery pack, based on the functional-morphological parameter adjustment cycle of the block building complex model, receives instructions from the deep deterministic policy gradient algorithm, and updates the state of the physical building complex model, batch outputs the functional-morphological parameters, and generates building complex layout schemes for business blocks, commercial blocks, residential blocks, public service blocks, and mixed blocks, that is, outputs the Rhinoceros model data of the 3D block building complex;

[0030] The building complex screening module is used to input the threshold range of building height, building density, and development intensity index for the target block form in the superior planning, connect the Rhinoceros model data of the 3D block building complex to the corresponding input port of the QueryModelObjects screening component, and adjust the screening conditions in the component's property panel. The screening conditions are the thresholds of building height, building density, and development intensity index for the target block form in the superior planning, run the definition processing data, and output the 3D block building complex model object set that meets the requirements of the superior planning after screening;

[0031] The display output module is used to superimpose the screened 3D block building complex models on the digital sand table of the target block; carry out scheme simulation and indicator display on the three-dimensional holographic digital sand table with voice recognition equipment and mixed reality wearable three-dimensional motion capture equipment, build a human-computer interaction instruction library, and determine the final scheme through mixed reality head-mounted devices in actual scenes for output.

[0032] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: the present invention improves the design efficiency while meeting the high coupling and coordination of function and form, effectively making up for the shortcomings of traditional research methods; the present invention applies the dual dimensions of function and form, and through the construction of the function-form grayscale co-occurrence matrix and the function-form association map, avoids the limitation of the single dimension of function or form, strengthens the coupling relationship between the function and form of the building complex, and makes the layout plan more scientific and credible; the present invention applies the deep deterministic policy gradient algorithm, combined with the Grasshopper battery pack to execute parameter input, to realize the generation of building complexes with dynamic interaction of function and form; by simulating different building complex design decisions and evaluating their effects, the nonlinear and multidimensional conditional constraints in the function-form layout of the solution are processed, the differentiated characteristics of the site itself are paid attention to, the pertinence and availability of the generated solution are enhanced, thereby creating an innovative and efficient building complex layout solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a flow chart of the present invention;

[0034] Figure 2 It is a diagram of the layout scheme iteration mechanism in the present invention;

[0035] Figure 3 It is a structural schematic diagram of the automatic layout system of building complex in the present invention. DETAILED DESCRIPTION

[0036] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0037] Example 1

[0038] like Figure 1 As shown, a method for automatic layout of building complexes based on function-form interaction comprises the following steps:

[0039] (1) Collect basic chassis data of the target block;

[0040] Based on the GIS geographic information platform, the design boundary data, road network data, ecological environment data and existing building data of the target block are collected, and data processing and spatial superposition are performed on the ArcGis platform to build a three-dimensional digital sand table;

[0041] The design boundary data of step (1) is provided by the local planning department. The road network data is retrieved from the Open Street Map (OSM) platform and exported by FME (Feature Manipulation Engine) after format conversion. The ecological environment data is obtained from the local ecological environment bureau. The ecological environment data includes mountain data, water data, ecologically sensitive area data and open space data. The existing building data is extracted from the Bing Maps open source data set. The existing building data includes latitude and longitude coordinates, building floor area and number of building floors.

[0042] (2) Constructing the gray-level co-occurrence matrix of land use function and spatial morphology;

[0043] The building area, surrounding road attributes, number of building floors and land use function data of several case blocks were collected, and the data were stored in a database management system (DBMS). All land use functions and spatial morphological index types of the building complexes in the case blocks were processed to form a function-morphology case library. All land use functions and spatial morphological index types in the function-morphology case library were uniformly displayed in grayscale grading to construct a grayscale co-occurrence matrix (GLCM) of land use functions and spatial morphology.

[0044] In step (2), the building area, number of floors, and surrounding road attributes are retrieved from the target block range on the Open Street Map platform and then converted to the format by FME (Feature Manipulation Engine); the land use function data is calculated by the TF-IDF algorithm using the POI data obtained from the Baidu Map Open Platform. The calculation formula is:

[0045]

[0046] Among them, v is the number of the block unit, w is a certain POI type, and k represents the t containing the POI to be retrieved. v The number of case blocks, S v,w represents the area of ​​type w land in v block unit, n v,w is the frequency of type w business in the vth block unit, D represents the total number of all POIs, |j:t v ∈d w | indicates the number of v contained; the corresponding relationship between POI types and land use functions is shown in Table 1.

[0047] Table 1 Correspondence between POI types and land use functions

[0048]

[0049] The types of spatial morphological indicators in step (2) include quantitative indicators of scale, symbolism and diversity; among them, scale indicators include total area, total base area, building density, floor area ratio and average height; symbolism indicators include maximum height and maximum base area; diversity indicators include height staggeredness and base area staggeredness; land use functions include production service functions, living service functions and public welfare service functions.

[0050] Based on the spatial position coordinates in the ArcGIS platform, the spatial morphological data of the building complex obtained from the OSM platform and the case block unit data obtained by road network buffering are spatially linked to obtain the base outlines and building storey information of all buildings in the case block, that is, all buildings in the same neighborhood share a neighborhood ID; then, according to the calculation formulas of various indicators, as shown in Table 2, the field calculator is used to calculate the scale index, iconic index and diversity index of the spatial morphology of each block.

[0051] Table 2 Calculation formulas for various indicators

[0052]

[0053]

[0054] In step (2), the gray-level symbiosis matrix of land use function and spatial morphology is constructed. The imread function in MATLAB (MatrixLaboratory) is used to input the spatial morphological data of the case block building complex reclassified by the natural break point method, including total area, total base area, building density, floor area ratio, average height, maximum height, maximum base area, height staggeredness, base area staggeredness, and land use function data including production service function, living service function, and public service function; the gray-level symbiosis matrix is ​​calculated by the graycomatrix function. The calculation formula of the gray-level symbiosis matrix correlation C is:

[0055]

[0056] Among them, N is the number of gray levels segmented by the natural break point method, P(p,q) is the value of each data of each case block building complex in the gray level co-occurrence matrix, p and q are the land use function of the case block building complex and the gray value of each data of each case block building complex respectively; the gray level co-occurrence matrix is ​​used to measure the correlation between the pixel and its adjacent pixels in the image, reflecting the regularity and consistency of its texture; based on the frequency of occurrence of pixel pairs with specific gray value combinations in specific direction and distance data, the correlation characteristics between various spatial morphological data and land use functions are quantitatively calculated;

[0057] (3) Generate a function-morphology association model based on the knowledge graph

[0058] Based on the arrangement rules between the pixels in the gray-level co-occurrence matrix, factor analysis is performed through the FactorAnalysis of the Python library to obtain the factor loading matrix file data, so as to reveal the potential connection and structure between the function-morphology indicators of the block building complex; the factor loading matrix file data in csv format is imported into the Neo4j graph database to complete the knowledge storage from text files to the graph database, provide graphic visualization function for the relationship between variables and factors, realize the construction of the knowledge graph of the function-morphology related factors of the block building complex, and generate the function-morphology association model based on the knowledge graph;

[0059] In step (3), a factor loading matrix is ​​constructed based on the spatial morphological factors and land use function type factors of the case block, and the spatial morphological data and land use function data are exported as factor data and variable data respectively; a FactorAnalysis object is created through the Python scikit-learn library, and the number of factors to be extracted is specified as 9 (n_components parameter) and the number of original variables is 1 (n_features parameter); the spatial morphological factor data of the case block is imported as a NumPy array as the data set to be analyzed, and the factor data is fitted using the fit_transform method, and the components_ attribute is used to obtain the factor loading matrix of the three major functional formats of production service function, life service function, and public welfare service function; where the size of the load represents the correlation strength between the building functional format variable and each factor, and the symbol represents the positive or negative relationship between them;

[0060] The correlation between factors in step (3) is based on the correlation structure of the factor loading matrix. The corresponding factor nodes, variable nodes and relationship networks are created through Neo4jDesktop. The Neo4j Browser graphic visualization tool is used to design the visual representation of the factor loading matrix, run Cypher queries to create a graphic view, and adjust the visualization settings to render and export the graphic data set.

[0061] (4) Iterate the layout scheme based on reinforcement learning algorithm

[0062] The Deep Deterministic Policy Gradient (DDPG) algorithm is used to generate building complexes with dynamic functional-morphological interactions. The functional-morphological factor load matrix of the block building complex in step (3) is imported and abstracted into the concepts of state, action, and reward.

[0063] In the DDPG deep deterministic policy gradient algorithm in step (4), the state is the current spatial morphological characteristics and land use function type of the block building complex, and the action is the next design decision made according to the current state, which is achieved by modifying the morphological attributes and functional attributes of the block building; the reward is a method used to evaluate the function-morphological matching degree of the block building through the reward function; based on the factor load values ​​of the corresponding functional business format and the arrangement rules between the pixels of the gray level co-occurrence matrix, the corresponding reward function is set to evaluate the pros and cons of the action;

[0064] The policy model output of the DDPG deep deterministic policy gradient algorithm is mapped to the control signal of the corresponding Grasshopper battery pack. Based on the function-morphology parameter adjustment loop of the block building complex model constructed in Rhino, the instructions from the function-morphology DDPG deep deterministic policy gradient descent algorithm are received, and the state of the physical building complex model is updated. The function-morphology parameters are batch-output to generate the building complex layout plans of business blocks, commercial blocks, residential blocks, public service blocks and mixed blocks, that is, the Rhinoceros model data of the 3D block building complex is output;

[0065] In the function-morphology parameter adjustment loop in step (4), the multi-agent and the environment chassis are first input to obtain the initial state S0; at the tth step, the spatial agent action A is randomly assigned t and B t , get state S t ; According to the reward function, S t Calculate the benefits and output the evaluation value Q of a series of action strategies in the current round t ; Change the current sequence (A t , B t , S t , Q t ) is stored in the experience pool D, and the experience replay and fixed Q network mechanism are combined to reduce process fitting, and the optimization training is used to obtain a series of action strategies with the largest Q value, such as Figure 2 As shown;

[0066] (5) Screening of building complex layout plans

[0067] Input the threshold range of building height, building density, and development intensity index for the target block form in the superior planning; connect the Rhinoceros model data of the 3D block building complex to the corresponding input port of the Query Model Objects screening component, and adjust the screening conditions in the component's property panel. The screening conditions are the thresholds of building height, building density, and development intensity index for the target block form in the superior planning. Run the definition to process the data and output the 3D block building complex model object set that meets the requirements of the superior planning after screening;

[0068] (6) Solution display, optimization and output

[0069] Based on the digital sand table of the target block constructed in step (1), the 3D models of the selected block building complex schemes are superimposed on the digital sand table respectively; a voice recognition device and a mixed reality wearable 3D motion capture device are installed on the 3D space holographic digital sand table to simulate the scheme and display indicators, and a human-computer interaction instruction library is constructed. The final scheme is determined and output through the mixed reality head-mounted device in the actual scene.

[0070] In terms of research objects, the present invention is not limited to existing cases, and can dynamically adjust design parameters according to specific needs and environmental feedback; in terms of research accuracy, the present invention is based on a reinforcement learning algorithm, and through continuous iteration and optimization of the design, the generated model covers various functional scenarios and design positioning; in terms of research conclusions, the present invention has dynamic adaptability and optimization capabilities, and can propose deeper design rules to deal with complex design problems restricted by multiple design goals and constraints.

[0071] Example 2

[0072] like Figure 3 As shown, based on the same inventive concept, the present invention discloses an automatic layout system for building complexes based on function-form interaction, comprising:

[0073] The digital sand table construction module is used to collect the design boundary data, road network data, ecological environment data and existing building data of the target block, perform data processing and spatial superposition, and construct a three-dimensional digital sand table; the digital sand table construction module specifically executes the specific steps in step (1) of Example 1;

[0074] A gray-level symbiotic matrix construction module is used to collect the building area, surrounding road attributes, number of building floors and land function data of several case blocks, process them to obtain all land functions and spatial morphological index types of the building complex in the case blocks, form a function-morphological case library, uniformly display all land functions and spatial morphological index types in grayscale classification, and construct a land function and spatial morphological gray-level symbiotic matrix; the gray-level symbiotic matrix construction module specifically performs the specific steps in step (2) in Example 1;

[0075] The function-morphology association model generation module is used to perform factor analysis based on the arrangement rules between each pixel point of the gray level co-occurrence matrix to obtain the function-morphology factor load matrix file data, import the factor load matrix file data into the graph database, complete the knowledge storage from the text file to the graph database, provide a graphical visualization function for the association relationship between variables and factors, and generate a function-morphology association model based on the knowledge graph; the function-morphology association model generation module specifically performs the specific steps in step (3) in Example 1;

[0076] The building complex layout scheme generation module is used to apply the deep deterministic policy gradient algorithm to generate a building complex with dynamic functional-morphological interaction, import the functional-morphological factor load matrix, and abstract it into the concepts of state, action and reward; the policy model output of the deep deterministic policy gradient algorithm corresponds to the control signal of the corresponding Grasshopper battery pack, based on the functional-morphological parameter adjustment cycle of the block building complex model, receives instructions from the deep deterministic policy gradient algorithm, and updates the state of the physical building complex model, batch outputs the functional-morphological parameters, and generates building complex layout schemes for business blocks, commercial blocks, residential blocks, public service blocks and mixed blocks, that is, outputs the Rhinoceros model data of the 3D block building complex; the building complex layout scheme generation module specifically executes the specific steps in step (4) of Example 1;

[0077] The building complex screening module is used to input the threshold range of building height, building density, and development intensity index for the target block form in the superior planning, connect the Rhinoceros model data of the 3D block building complex to the corresponding input port of the QueryModelObjects screening component, and adjust the screening conditions in the component's property panel. The screening conditions are the thresholds of building height, building density, and development intensity index for the target block form in the superior planning, run the definition processing data, and output the 3D block building complex model object set that meets the requirements of the superior planning after screening;

[0078] The display output module is used to superimpose the screened 3D block building complex models on the digital sand table of the target block; carry out scheme simulation and indicator display on the three-dimensional holographic digital sand table with voice recognition equipment and mixed reality wearable three-dimensional motion capture equipment, build a human-computer interaction instruction library, and determine the final scheme through mixed reality head-mounted devices in actual scenes for output.

Claims

1. A method for automatic layout of building complexes based on function-form interaction, characterized in that: The steps include: (1) Collect the design boundary data, road network data, ecological environment data and existing building data of the target block, perform data processing and spatial superposition, and construct a three-dimensional digital sand table; (2) Collect data on building area, surrounding road attributes, number of building floors and land use functions of several case blocks, process them to obtain all land use functions and spatial morphological index types of the building complexes in the case blocks, form a function-morphological case library, uniformly display all land use functions and spatial morphological index types in grayscale classification, and construct a grayscale symbiosis matrix of land use functions and spatial morphology; (3) Based on the arrangement rules between the pixels of the gray-level co-occurrence matrix, factor analysis is performed to obtain the function-morphology factor loading matrix file data, and the factor loading matrix file data is imported into the graph database to complete the knowledge storage from the text file to the graph database. The relationship between variables and factors is provided with a graphical visualization function, and a function-morphology association model based on the knowledge graph is generated; (4) Applying the deep deterministic policy gradient algorithm to generate building complexes with dynamic interaction between function and form, importing the function-form factor loading matrix in step (3), and abstracting it into the concepts of state, action, and reward; The policy model output of the deep deterministic policy gradient algorithm is mapped to the control signal of the corresponding Grasshopper battery pack. Based on the function-morphology parameter adjustment cycle of the block building complex model, the command from the deep deterministic policy gradient algorithm is received, and the state of the physical building complex model is updated. The function-morphology parameters are batch output, and the building complex layout plans for business blocks, commercial blocks, residential blocks, public service blocks and mixed blocks are generated respectively, that is, the Rhinoceros model data of the 3D block building complex is output; (5) Input the threshold range of building height, building density, and development intensity index for the target block form in the superior planning, connect the Rhinoceros model data of the 3D block building complex to the corresponding input port of the Query Model Objects screening component, and adjust the screening conditions in the component's property panel. The screening conditions are the threshold range of building height, building density, and development intensity index for the target block form in the superior planning. Run the definition to process the data and output the 3D block building complex model object set that meets the requirements of the superior planning after screening; (6) Based on the target block digital sand table constructed in step (1), superimposing the screened 3D block building complex model on the digital sand table; Equipped with voice recognition equipment and mixed reality wearable 3D motion capture equipment on the 3D holographic digital sandbox, solutions are simulated and indicators are displayed. A human-computer interaction instruction library is built, and the final solution is determined and output through mixed reality head-mounted devices in actual scenarios.

2. The automatic layout method of building complex based on function-form interaction according to claim 1, characterized in that: In the step (1), based on the GIS geographic information platform, the design boundary data, road network data, ecological environment data and existing building data of the target block are collected, and data processing and spatial superposition are performed in the ArcGis platform to construct a three-dimensional digital sand table; the ecological environment data includes mountain data, water body data, ecologically sensitive area data and open space data, and the existing building data includes latitude and longitude coordinates, building floor area and building number of floors information.

3. The automatic layout method of building complex based on function-form interaction according to claim 1 is characterized by: In the step (2), the building area, number of floors, and surrounding road attributes are retrieved from the target block range on the Open Street Map platform and then converted into a new format; the land use function data is calculated from the POI data using the TF-IDF algorithm, and the calculation formula is: Among them, v is the number of the block unit, w is a certain POI type, and k represents the t containing the POI to be retrieved. v The number of case blocks, S v,w represents the area of ​​type w land in the v block unit, n v,w is the frequency of type w business in the vth block unit, D represents the total number of all POIs, |j:t v ∈d w | indicates the number of v contained; there is a corresponding relationship between POI types and land use functions.

4. The automatic layout method of building complex based on function-form interaction according to claim 3 is characterized by: The types of spatial morphological indicators in step (2) include quantitative indicators of scale, symbolism and diversity; among which scale indicators include total area, total base area, building density, floor area ratio and average height; symbolism indicators include maximum height and maximum base area; diversity indicators include height staggeredness and base area staggeredness; land use functions include production service functions, living service functions and public welfare service functions.

5. The automatic layout method of building complex based on function-form interaction according to claim 4 is characterized by: The specific steps of processing in step (2) to obtain all land use functions and spatial morphological index types of the case block building complex are as follows: based on the spatial position coordinates, in the ArcGIS platform, the building complex spatial morphological data obtained from the OSM platform and the case block unit data obtained by road network buffering are spatially linked to obtain all building base outlines and building floor information in the case block, that is, all buildings in the same neighborhood share a neighborhood ID; and then according to the calculation formulas of various indicators, the scale index, symbolic index and diversity index of the spatial morphology of each block are calculated.

6. The automatic layout method of building complex based on function-form interaction according to claim 5 is characterized by: In the step (2), the gray-level co-occurrence matrix of land use function and spatial morphology is constructed by using the imread function in MATLAB to input the spatial morphology data and land use function data of the case block building complex reclassified by the natural break point method; the gray-level co-occurrence matrix is ​​calculated by the graycomatrix function, and the calculation formula of the gray-level co-occurrence matrix correlation C is: Among them, N is the number of gray levels segmented by the natural break point method, P(p,q) is the value of each data of each case block building complex in the gray level co-occurrence matrix, p and q are the land use function of the case block building complex and the gray value of each data of each case block building complex, respectively; the gray level co-occurrence matrix is ​​used to measure the correlation between the pixel and its adjacent pixels in the image, reflecting the regularity and consistency of its texture; based on the frequency of occurrence of pixel pairs with specific gray value combinations in specific direction and distance data, the correlation characteristics between various spatial morphological data and land use functions are quantitatively calculated.

7. The automatic layout method of building complex based on function-form interaction according to claim 1 is characterized by: In the step (3), a factor loading matrix is ​​constructed based on the spatial morphological factors and land use function type factors of the case block, and the spatial morphological data and land use function data are respectively exported as factor data and variable data; a FactorAnalysis object is created through the scikit-learn library of Python, and the number of factors to be extracted is specified as 9 and the number of original variables is specified as 1; the spatial morphological factor data of the case block is imported as a NumPy array as the data set to be analyzed, the factor data is fitted using the fit_transform method, and the factor loading matrix of the three major functional business types of production service function, life service function, and public welfare service function is obtained using the components_ attribute; wherein the size of the load represents the correlation strength between the building functional business type variable and each factor, and the symbol represents the positive or negative relationship between them; the correlation relationship between the factors in the step (3) is based on the correlation structure of the factor loading matrix, and the corresponding factor nodes, variable nodes and relationship networks are created through Neo4j Desktop; the Neo4j Browser graphic visualization tool is used to design a visual representation of the factor loading matrix, a Cypher query is run to create a graphic view, and the visualization settings are adjusted to render and export the graphic data set.

8. The automatic layout method of building complex based on function-form interaction according to claim 1 is characterized by: In the deep deterministic policy gradient algorithm in step (4), the state is the current spatial morphological characteristics and land use function type of the block building complex; the action is the next design decision made according to the current state, which is achieved by modifying the morphological attributes and functional attributes of the block building; the reward is a method for evaluating the function-morphological matching degree of the block building through a reward function; based on the factor load values ​​of the corresponding functional business type and the arrangement rules between the pixels of the gray level co-occurrence matrix, a corresponding reward function is set to evaluate the pros and cons of the action.

9. The automatic layout method of building complex based on function-form interaction according to claim 8 is characterized by: The function-morphology parameter adjustment loop in step (4) first inputs the multi-agent and the environment chassis to obtain the initialization state S0; at the tth step, randomly assigns the spatial agent actions Bt and Bt to obtain the state St; calculates the benefit of St according to the reward function, and outputs the evaluation value Qt of a series of action strategies in the current round; stores the current sequence (At, Bt, St, Qt) in the experience pool D, combines the experience replay and the fixed Q network mechanism to reduce process fitting, and optimizes the training to obtain a series of action strategies with the maximum Q value.

10. An automatic building layout system based on function-form interaction, characterized in that: include: The digital sand table construction module is used to collect the design boundary data, road network data, ecological environment data and existing building data of the target block, perform data processing and spatial superposition, and construct a three-dimensional digital sand table; The gray-level co-occurrence matrix construction module is used to collect the building area, surrounding road attributes, number of building floors and land use function data of several case blocks, and process them to obtain all the land use functions and spatial morphological index types of the building complex in the case blocks, forming a function-morphological case library, and uniformly displaying all the land use functions and spatial morphological index types in gray-level classification to construct the land use function and spatial morphological gray-level co-occurrence matrix; The function-morphology association model generation module is used to perform factor analysis based on the arrangement rules between the pixels of the gray-level co-occurrence matrix to obtain the function-morphology factor load matrix file data, import the factor load matrix file data into the graph database, complete the knowledge storage from the text file to the graph database, provide a graphical visualization function for the association between variables and factors, and generate a function-morphology association model based on the knowledge graph; Building complex layout scheme generation module, which is used to apply deep deterministic policy gradient algorithm to generate building complexes with dynamic interaction between function and form, import the function-form factor load matrix, and abstract it into the concepts of state, action and reward; The policy model output of the deep deterministic policy gradient algorithm is mapped to the control signal of the corresponding Grasshopper battery pack. Based on the function-morphology parameter adjustment cycle of the block building complex model, the command from the deep deterministic policy gradient algorithm is received, and the state of the physical building complex model is updated. The function-morphology parameters are batch output, and the building complex layout plans for business blocks, commercial blocks, residential blocks, public service blocks and mixed blocks are generated respectively, that is, the Rhinoceros model data of the 3D block building complex is output; The building complex screening module is used to input the threshold range of building height, building density, and development intensity index for the target block form in the superior planning, connect the Rhinoceros model data of the 3D block building complex to the corresponding input port of the Query Model Objects screening component, and adjust the screening conditions in the component's property panel. The screening conditions are the thresholds of building height, building density, and development intensity index for the target block form in the superior planning, run the definition processing data, and output the 3D block building complex model object set that meets the requirements of the superior planning after screening; The display output module is used to overlay the 3D block building complex model that has passed the screening on the digital sand table according to the target block digital sand table; Equipped with voice recognition equipment and mixed reality wearable 3D motion capture equipment on the 3D holographic digital sandbox, solutions are simulated and indicators are displayed. A human-computer interaction instruction library is built, and the final solution is determined and output through mixed reality head-mounted devices in actual scenarios.

Images