High-precision solar radiation analysis modeling method based on point cloud

Abstract

This application discloses a high-precision solar radiation analysis modeling method based on point clouds. The method includes: extracting and storing feature point information; and automatically constructing the solar radiation analysis model. Advantages: It effectively solves the problem of rapidly constructing high-precision solar radiation analysis models based on feature point information extracted from point cloud data. Compared with traditional modeling methods, it significantly reduces the workload and difficulty of field measurement and indoor modeling. By extracting feature points from point cloud data and storing them according to rules, the model can be directly generated, eliminating the need for traditional modeling methods such as drawing building outlines, calculating the height of various parts of the building, and manually inputting parameters. The requirements for the professional skills of the operators are also greatly reduced. While ensuring the accuracy of the results, it greatly improves the efficiency of solar radiation analysis modeling, providing an efficient technical solution for establishing solar radiation analysis models of existing main and secondary buildings.

Description

Technical Field

[0001] This invention relates to the field of engineering surveying, and in particular to a method for acquiring point cloud data based on laser point cloud technology, extracting key feature points of buildings, and constructing a high-precision solar radiation analysis model based on point clouds. Background Technology

[0002] Current methods for solar radiation analysis, primarily using modeling techniques for both the primary and secondary buildings, mainly employ a fully field-based digital mapping approach to measure the building's planar location and height. This involves simultaneously taking photographs and creating sketches on-site. Subsequently, based on the data obtained from the field measurements, along with the on-site photographs and sketches, three-dimensional models of the primary and secondary buildings for solar radiation analysis are created in the office. This method has significant drawbacks and limitations.

[0003] Surveying methods based on full-field digital mapping suffer from drawbacks such as low operational efficiency, high operational difficulty, high labor intensity, high project costs, long surveying cycles, and high requirements for the professional skills of the personnel. Furthermore, significant problems arise in the subsequent indoor model-building phase, particularly when the existing building structure is complex, leading to high communication costs and potential quality issues due to misunderstandings.

[0004] The shortcomings and deficiencies of full-field digital mapping result in a large investment of manpower and resources, as well as high time costs. How to improve operational efficiency, reduce project costs, and shorten project cycles while ensuring product accuracy has become a problem that needs to be solved.

[0005] With the increasing maturity of laser scanning technology, its application in engineering surveying has become more widespread, providing a new approach for rapid and efficient data acquisition. However, how to quickly construct a solar radiation analysis model based on massive point cloud data has become an urgent problem for researchers in this field.

[0006] Simultaneously, a method is needed to rapidly construct high-precision solar radiation analysis models based on feature point information extracted from point cloud data. Compared with traditional modeling methods, this approach simplifies, visualizes, and automates the process, significantly reducing the workload and difficulty of field surveying and indoor modeling. By extracting feature points from point cloud data and storing them according to rules, models can be directly generated, eliminating the need for traditional modeling methods that require drawing building outlines, calculating the height of various building parts, and manually inputting parameters. The requirements for the professional skills of the operators are also greatly reduced. While ensuring the accuracy of the results, this approach significantly improves the efficiency of solar radiation analysis modeling, providing an efficient technical solution for establishing solar radiation analysis models of existing main and secondary buildings. Summary of the Invention

[0007] This invention provides a high-precision solar radiation analysis modeling method based on point clouds. This method effectively solves the problem of rapidly constructing high-precision solar radiation analysis models based on feature point information extracted from point cloud data. Compared with traditional modeling methods, the process is simplified, visualized, and intelligent, significantly reducing the workload and difficulty of field measurement and indoor modeling. By extracting feature points from point cloud data and storing them according to rules, the model can be directly generated, eliminating the need for traditional modeling methods such as drawing building outlines, calculating the height of various building parts, and manually inputting parameters. The requirements for the professional skills of the operators are also greatly reduced. While ensuring the accuracy of the results, it greatly improves the efficiency of solar radiation analysis modeling, providing an efficient technical solution for establishing solar radiation analysis models of existing main and secondary buildings.

[0008] This invention provides a method for modeling a high-precision solar radiation analysis model based on point clouds, wherein the method includes the following steps:

[0009] Extraction and storage of feature point information: A 3D laser scanner is used for field data acquisition to obtain building point cloud data; the scanned point cloud is processed by point cloud data registration, coordinate system transformation, noise reduction and thinning; feature point information of buildings / structures is extracted according to rules and categories, the feature point number and coordinate information (x, y, z) are exported, and feature point files of the same type are named according to the same rules or stored in the same file path;

[0010] Automatically construct a solar radiation analysis model: Traverse the feature point files, and based on the feature point type, feature point number, coordinate value, and remarks, automatically construct the building main body, auxiliary structure model, solar window model, and roof model corresponding to each feature point file. Each individual model is placed in the corresponding three-dimensional space according to the coordinate value, and automatically combined to generate a complete high-precision solar radiation analysis model.

[0011] A method for modeling a high-precision solar radiation analysis model based on point clouds, wherein the extraction and storage of feature point information includes the following steps:

[0012] Feature point classification, extraction, numbering, and annotation: Extract feature points from the point cloud data for the main building, auxiliary structures, sunlight windows, and rooftops, and collect feature point coordinate information;

[0013] Feature point information storage: The feature point information of the main building and auxiliary structures with different bottom elevations and main body heights, the feature point information of windows belonging to different floors, and the feature point information of slopes that do not belong to the same slope are extracted by type and stored in feature point files according to rules. Among them, the feature points of the main building belonging to the same block are saved in a separate feature point file, the feature points of the auxiliary structures belonging to the same block are saved in a separate feature point file, the feature points of windows with the same window sill elevation and window height on the same floor are saved in a separate feature point file, and the feature points of the same slope are saved in a separate feature point file.

[0014] A high-precision solar radiation analysis modeling method based on point clouds, wherein the feature point classification, extraction, numbering, and annotation information include the following steps:

[0015] Exporting the feature point extraction model: Define sample input functions based on feature points of the main building, auxiliary structures, sunlight windows, and rooftop, which are used to train and evaluate the feature point extraction model; define hook callback functions and register them in the estimator framework to output detailed information during the training process; establish a for loop and perform training and evaluation operations on the feature point extraction model inside the loop, while outputting relevant information; after training is completed, export the feature point extraction model.

[0016] Feature point extraction, numbering, and remarks: Based on the feature point extraction model, feature points constituting the main building structure, auxiliary structures, windows, and rooftops are automatically extracted through calculation. The automatically extracted feature points are checked, and erroneous feature points are corrected through human-computer interaction. For feature points belonging to the same block constituting the main building structure or auxiliary structures, window features on the same floor, and feature points constituting the same rooftop, a starting point is randomly determined, and then the feature points are numbered sequentially. Remarks are added as needed; otherwise, they are not required. The formula for automatically extracting feature points constituting the main building structure, auxiliary structures, windows, and rooftops is:

[0017]

[0018] In the formula:

[0019] U(s,a) Feature point extraction probability

[0020] c puct exploration level constant

[0021] P(s,a) Prior probability

[0022] N(s,b) is the number of times the parent node has been visited.

[0023] N(s,a) represents the number of visits to a certain child node.

[0024] When extracting feature point information that constitutes the main building, auxiliary structures, and sunlight windows, at least one feature point at the bottom edge and one feature point at the top edge should be ensured.

[0025] A method for building a high-precision solar radiation analysis model based on point clouds, wherein automatically constructing the solar radiation analysis model includes the following steps:

[0026] Automatically construct building main structure, auxiliary structure, and window models: Retrieve all feature point files of the building main structure, auxiliary structure, and windows stored in a specified path and named according to a specific naming method, and read each feature point file one by one; traverse the feature point data of each feature point file line by line, and store the feature point data of each feature point file into different arrays; calculate the bottom elevation and main body height of each building main structure and auxiliary structure, and the window sill elevation and window height of each floor / each window based on the z-coordinate values ​​of the feature points in different arrays; connect the feature points in sequence using contour lines based on the (x, y) coordinate values ​​of the feature points belonging to the same block extracted from the array, generate the outer contour line of the building main structure or the outer contour line of the auxiliary structure, and establish the building main structure and auxiliary structure models based on the bottom elevation and main body height of the contour lines; extract the (x, y) coordinate values ​​of the left and right window sill feature points of each window on the same floor from the array, and establish the window model using a two-point window insertion method based on the window sill feature point coordinate values, window sill elevation, and window height;

[0027] Automatically construct the slope top model: retrieve all feature point files of the slope top stored in the specified path and named according to a specific naming method, and read each feature point file one by one; then traverse the feature point data in the feature point file line by line, and store the feature point data of each slope top feature point file into different arrays; generate a three-dimensional mesh surface by connecting the slope top feature points in sequence to construct the slope top model.

[0028] A high-precision solar radiation analysis modeling method based on point clouds, wherein the automatic construction of the building main body, auxiliary structures, and solar window models includes the following steps:

[0029] Retrieve feature point files for the main building, ancillary structures, and sun windows: Retrieve and obtain all feature point files for the main building, ancillary structures, and sun windows, and iterate through and read the feature point files for the main building, ancillary structures, and sun windows one by one;

[0030] Read feature point coordinates and store them in an array: Iterate through the feature point data in each feature point file, obtain the number and coordinate information (x, y, z) of each feature point in each feature point file, and determine whether coordinate unit conversion is required. If it is determined that coordinate unit conversion is required, after iterating through and obtaining the feature point coordinate information, convert the coordinate units according to the formula. After conversion, store the number and coordinate information of feature points belonging to the main body and auxiliary structures of the same block, and feature points belonging to each sun window on the same floor, in the corresponding arrays in order. One feature point file corresponds to one array.

[0031] The formula for converting coordinate units is:

[0032] x = n × x0

[0033] y = n × y0

[0034] z = n × z0

[0035] In the formula:

[0036] x0, y0, and z0 represent the coordinate values ​​of feature points extracted from the point cloud;

[0037] n represents the scaling factor;

[0038] x, y, and z represent the coordinate values ​​corresponding to the plotted solar radiation model;

[0039] When the original coordinate values ​​of the point cloud data are in meters, and the coordinate values ​​used to draw the solar radiation model are in millimeters, then n is set to 1000 for coordinate unit conversion.

[0040] When it is determined that no coordinate unit conversion is required, the feature point number and coordinate value information (x, y, z) of the feature point file are obtained, and the feature points belonging to the same block of the main building and auxiliary structure, and the feature points belonging to each sun window on the same floor are directly stored in the same array in order.

[0041] Calculate the base elevation: Extract the z-coordinate values ​​of feature points belonging to the same building block and its auxiliary structures, as well as feature points of windows with the same sill elevation and window height, from the array, and add them to their corresponding sequences; traverse the sequence of each feature point belonging to the same building block and its auxiliary structures, as well as each feature point of windows with the same sill elevation and window height, to calculate the base elevation of each different building block and its auxiliary structures, and the sill elevation of the windows; the specific method is as follows:

[0042] Iterate through the feature points of the main building and auxiliary structures belonging to the same block, as well as the feature points of windows with the same sill elevation and window height, and add them to their respective sequences. Sort the z-coordinate values ​​of the feature points in the sequences, set a threshold Δh, with the threshold range between 0m and 1m, and calculate the average z-coordinate of the bottom feature points of all the bottom feature points of the main building and auxiliary structures belonging to the same block, as well as the bottom feature points of windows with the same sill elevation and window height within the threshold range. Calculate the bottom elevation h of the windows of different blocks, the main building and auxiliary structures, and different floors. 底标高 The specific calculation process is as follows:

[0043] First, determine the coordinates of the lowest point:

[0044] z min =min(z) i (i = 1, 2, ..., n)

[0045] Secondly, based on the set threshold Δh, feature points participating in the calculation are selected. If the j-th feature point z j If the difference between the coordinates of the lowest point and the coordinates of the lowest point is within the threshold range, then it participates in the next step h. 底标高 The calculation determines whether the feature points participate in h. 底标高 The formula for the calculation is:

[0046] z j -z min ≤Δh(j=1,2,…,m)

[0047] Finally, calculate the bottom elevation h. 底标高 :

[0048]

[0049] When the threshold Δh is 0, then h 底标高 The calculation formula is:

[0050] h 底标高 =min(z) i (i = 1, 2, ..., n)

[0051] In the formula:

[0052] h 底标高 This refers to the bottom elevation of the main building structure or its ancillary structures, or the windowsill elevation of a window that receives sunlight.

[0053] z i Let z be the z-coordinate of the i-th point, min be the minimum value, and n be the total number of feature points;

[0054] z j This represents the z-coordinate value of the j-th bottom feature point that is within the threshold range and participates in the calculation; m represents the number of bottom feature points participating in the calculation.

[0055] Calculate the main building height and window height: Extract the z-coordinate values ​​of feature points belonging to the same building block and its auxiliary structures, as well as feature points of windows with the same sill elevation and window height, from the array, and add them to their corresponding sequences; traverse the sequence of each feature point belonging to the same building block and its auxiliary structures, as well as each feature point of windows with the same sill elevation and window height, and calculate the main building height and window height of each different building block and its auxiliary structures; the specific method is as follows:

[0056] Iterate through the feature points of the main building and auxiliary structures belonging to the same block, as well as the feature points of windows with the same sill elevation and window height, using their z-coordinates. Add these z-coordinates to their respective sequences, sort the z-coordinate values ​​of the feature points in the sequences, and set a threshold Δh, with the threshold range between 0m and 1m. Calculate the average z-coordinates of the top feature points of all the top feature points of the main building and auxiliary structures belonging to the same block, as well as the top feature points of windows with the same sill elevation and window height within the threshold range. Calculate the top height h of the windows for different blocks of the main building, auxiliary structures, and different floors. 顶部高 Determine the main body height and window height h. 主体高 for h 顶部高 with h 底标高 The difference is calculated as follows:

[0057] First, determine the coordinates of the highest point:

[0058] z max =max(z) i (i = 1, 2, ..., n)

[0059] Secondly, based on the set threshold Δh, feature points participating in the calculation are selected. If the j-th feature point z j If the difference between the coordinates of the highest point and the coordinates of the highest point is within the threshold range, then it participates in the next step h. 顶部高 The calculation determines whether the feature points participate in h. 顶部高 The formula for the calculation is:

[0060] z max -z j ≤Δh(i=1,2,…,m)

[0061] Then, calculate the top height h. 顶部高 :

[0062]

[0063] When the threshold Δh is 0, then h 顶部高 The calculation formula is:

[0064] h 顶部高 =max(z)i (i = 1, 2, ..., n)

[0065] Finally, calculate the main body height h. 主体高 :

[0066] h 主体高 =h 顶部高 -h 底标高

[0067] In the formula:

[0068] h 主体高 The height of the main building and its ancillary structures, and the height of the windows for sunlight; h 顶部高 h is the height of the top of the main building, auxiliary structures, and windows. 底标高 This refers to the bottom elevation of the main building structure or its ancillary structures, or the windowsill elevation of a window that receives sunlight.

[0069] z i Let z be the z-coordinate of the i-th feature point, max be the maximum value, and n be the total number of feature points;

[0070] z j This represents the z-coordinate value of the j-th top feature point that is within the threshold range and participates in the calculation; m represents the number of top feature points participating in the calculation.

[0071] Determine the modeling type: Determine the model type according to the main building, auxiliary structures, and sunlight windows, and construct the model based on the determined model type;

[0072] Building Main Structure and Ancillary Structure Model Construction: Based on the feature point coordinate information extracted from the array, the bottom elevation (h) of the main structure and ancillary structures is obtained from the bottom elevation calculation step. 底标高 The numerical values ​​for calculating the main building height and window height are obtained from the calculation steps. 主体高 The system uses numerical values ​​to draw building outlines and automatically constructs building models corresponding to feature point files of each building and its ancillary structures.

[0073] Sunlight window model construction: Based on the coordinate values ​​of feature points of each sunlight window on the same floor extracted from the array, the window sill elevation (h) obtained in the bottom elevation step is calculated. 底标高 The window height (h) is obtained by calculating the main body height and window height. 主体高 The numerical values ​​are used to extract the windowsill points of the sunshine windows, and one or more sunshine window models corresponding to each sunshine window feature point file are automatically constructed.

[0074] A high-precision solar radiation analysis modeling method based on point clouds, wherein the construction of the building main body and auxiliary structure models includes the following steps:

[0075] Drawing building outlines: Extract the (x, y) coordinates of each feature point belonging to the same block of the main building and its auxiliary structures from the array; automatically connect the feature points with outlines according to the coordinate order to generate the outer outline of the main building or the auxiliary structures; determine whether the outline is closed according to requirements; when it is determined that the outline needs to be closed, connect the last feature point to the first feature point end to end and set it as a closed outline; when it is determined that the outline does not need to be closed, the last feature point does not need to be connected to the first feature point.

[0076] Model construction: Set the bottom elevation value of the contour line to the bottom elevation (h) obtained from the bottom elevation calculation step. 底标高 The thickness value of the outline is set to the main body height (h) obtained from the steps of calculating the main body height and window height. 主体高 Set the bottom elevation value of the outline to the bottom elevation (h) obtained from the bottom elevation calculation step. 底标高 After that, the block is modeled by stretching it up according to the outline. The height of the stretched block is the main body height (h) obtained from the steps of calculating the main body height and window height. 主体高 );

[0077] When the outer contour shape of balconies on each floor is consistent and the spacing between balconies on different floors is within the tolerance range, the following method is adopted: first, a balcony model of a certain floor is established based on the contour line, bottom elevation, and main structure height; then, based on the feature point file information, the spacing value △H of the balconies in the Z direction is calculated, and the balconies of each floor are copied according to the spacing value array.

[0078] The calculation formula is:

[0079]

[0080] In the formula:

[0081] z m z m+1 This indicates the bottom elevation or top elevation of the balcony on the m-th or m+1-th floor;

[0082] n represents the number of balcony floors involved in the calculation and modeling;

[0083] △H represents the average spacing value.

[0084] A method for modeling a high-precision solar radiation analysis model based on point clouds, wherein the construction of the solar radiation window model includes the following steps:

[0085] Extracting left and right window sill points of sunlit windows: Extract the (x,y) coordinate values ​​of feature points of each sunlit window on the same floor from the array to distinguish between left and right window sill points;

[0086] Insert windows: Using the left and right endpoints of the window as the start and end points, insert windows at these two points to provide sunlight. Set the windowsill elevation to the windowsill elevation (h) obtained in the calculation of the bottom elevation step. 底标高 Set the window height to the height of the main body and the window height obtained in the window height calculation step (h). 主体高 );

[0087] A high-precision solar radiation analysis modeling method based on point clouds, wherein the automatic construction of the slope top model includes the following steps:

[0088] Retrieve slope crest feature point files: Based on the file name and storage path of the feature point files, retrieve all slope crest feature point files, and then iterate through and read the slope crest feature point files one by one;

[0089] Read feature point coordinates and store them in an array: Iterate through the feature point data in each feature point file line by line, obtain the coordinates (x, y, z) and remarks of each feature point in each feature point file, and store the coordinates and remarks of the feature points that constitute the top of the slope in the corresponding array in order; Perform coordinate unit conversion according to the steps of reading feature point coordinates and storing them in an array in the automatic construction of building main body, auxiliary structure and sunlight window model;

[0090] To create a 3D mesh surface, connect all feature points on the same slope point by point. The specific method is as follows: traverse the feature points in the array and extract them, connect them to construct a 3D surface, and calculate the slope and aspect of the 3D surface; calculate and obtain 3D surfaces with the same slope and aspect, extract the points that make up the 3D surface, and store their (x, y, z) values ​​in an array; iterate through the calculations, storing points on the same slope surface in the same array, resulting in a separate array for each slope surface; remove duplicate coordinates of the feature points (x, y, z) in the slope coordinate array; calculate the order of the coordinate points in each array, and connect the feature points on the slope surface with the same slope point one by one according to the order to generate a 3D mesh surface.

[0091] A high-precision solar radiation analysis modeling method based on point clouds, wherein: the ancillary structures include: balconies, eaves, columns, air conditioning partitions, unit partitions, moldings, sunshades, parapet walls of the roof, stairwells leading to the roof, water tank rooms, decorative components, and elevator machine rooms; the feature point information is stored in the following way: feature point information of the main building and ancillary structures with different bottom elevations and main heights, feature point information of solar windows belonging to different floors, and feature point information of slopes not belonging to the same slope are saved to different feature point files, and the feature point files are named according to the feature point type. Different feature point files are distinguished by adding suffixes, etc. Feature point information of building main body and auxiliary structure with different bottom elevations and main body heights, feature point information of sun windows belonging to different floors, and feature point information of slopes not belonging to the same slope are stored in the same file path according to category, and the folder is named according to the feature point type. The method of building the model based on the bottom elevation and main body height of the outline includes: modeling by setting the bottom elevation and main body height of the outline; or modeling by setting the bottom elevation of the outline and then raising the block, and the height value of the raised block is the main body height value.

[0092] Therefore, it can be seen that:

[0093] The high-precision solar radiation analysis modeling method based on point clouds in this invention effectively solves the problem of rapidly constructing high-precision solar radiation analysis models based on feature point information extracted from point cloud data. Compared with traditional modeling methods, the operation process is simplified, visualized, and intelligent, significantly reducing the workload and difficulty of field measurement and indoor modeling. By extracting feature points from point cloud data and storing them according to rules, the model can be directly generated, eliminating the need for traditional modeling methods to draw building outlines, calculate the height of various parts of the building, and manually input parameters for modeling. The requirements for the professional skills of the operators are also greatly reduced. While ensuring the accuracy of the results, the efficiency of solar radiation analysis modeling is greatly improved, providing an efficient technical solution for establishing solar radiation analysis models of existing main and secondary buildings. Attached Figure Description

[0094] Figure 1 A schematic diagram of the overall process of a high-precision solar radiation analysis modeling method based on point clouds provided in an embodiment of the present invention;

[0095] Figure 2 A flowchart illustrating the steps of extracting and storing feature point information in a point cloud-based high-precision solar radiation analysis modeling method provided in an embodiment of the present invention;

[0096] Figure 3 A flowchart illustrating the steps of feature point classification, extraction, numbering, and annotation in the high-precision solar radiation analysis modeling method based on point cloud provided in the embodiments of the present invention;

[0097] Figure 4 A flowchart illustrating the steps of automatically constructing a solar radiation analysis model in the point cloud-based high-precision solar radiation analysis model modeling method provided in the embodiments of the present invention;

[0098] Figure 5 A flowchart illustrating the steps of automatically constructing the main building, auxiliary structures, and sunlight window models in the point cloud-based high-precision solar analysis model modeling method provided in the embodiments of the present invention;

[0099] Figure 6 A flowchart illustrating the steps of building main body and auxiliary structure model construction in the point cloud-based high-precision solar radiation analysis model modeling method provided in the embodiments of the present invention;

[0100] Figure 7 A flowchart illustrating the steps of constructing a solar window model in a point cloud-based high-precision solar analysis model modeling method provided in an embodiment of the present invention;

[0101] Figure 8 A flowchart illustrating the steps of automatically constructing a slope top model in a point cloud-based high-precision solar radiation analysis model modeling method provided in an embodiment of the present invention;

[0102] Figure 9 This is one of the implementation effect diagrams of the high-precision solar radiation analysis modeling method based on point clouds provided in the embodiments of the present invention;

[0103] Figure 10 This is one of the implementation effect diagrams of the high-precision solar radiation analysis modeling method based on point clouds provided in the embodiments of the present invention.

[0104] Figure 11 This is one of the implementation effect diagrams of the high-precision solar radiation analysis modeling method based on point clouds provided in the embodiments of the present invention.

[0105] Figure 12 This is one of the implementation effect diagrams of the high-precision solar radiation analysis modeling method based on point clouds provided in the embodiments of the present invention.

[0106] Figure 13 This is one of the implementation effect diagrams of the high-precision solar radiation analysis modeling method based on point clouds provided in the embodiments of the present invention.

[0107] Figure 14 This is one of the implementation effect diagrams of the high-precision solar radiation analysis modeling method based on point clouds provided in the embodiments of the present invention.

[0108] Figure 15 This is one of the implementation effect diagrams of the high-precision solar radiation analysis modeling method based on point clouds provided in the embodiments of the present invention.

[0109] Figure 16 This is one of the implementation effect diagrams of the high-precision solar radiation analysis modeling method based on point clouds provided in the embodiments of the present invention. Detailed Implementation

[0110] To enable those skilled in the art to better understand the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention, but are not intended to limit the present invention.

[0111] Example 1:

[0112] Figure 1 A modeling method for high-precision solar radiation analysis based on point clouds, such as Figure 1 As shown, the method includes the following steps:

[0113] Extraction and storage of feature point information: A 3D laser scanner is used for field data acquisition to obtain building point cloud data; the scanned point cloud is processed by point cloud data registration, coordinate system transformation, noise reduction and thinning; feature point information of buildings / structures is extracted according to rules and categories, the feature point number and coordinate information (x, y, z) are exported, and feature point files of the same type are named according to the same rules or stored in the same file path;

[0114] Automatically construct a solar radiation analysis model: Traverse the feature point files, and based on the feature point type, feature point number, coordinate value, and remarks, automatically construct the building main body, auxiliary structure model, solar window model, and roof model corresponding to each feature point file. Each individual model is placed in the corresponding three-dimensional space according to the coordinate value, and automatically combined to generate a complete high-precision solar radiation analysis model.

[0115] like Figure 2 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for extracting and storing feature point information:

[0116] Feature point classification, extraction, numbering, and annotation: Extract feature points from the point cloud data for the main building, auxiliary structures, sunlight windows, and rooftops, and collect feature point coordinate information;

[0117] Feature point information storage: The feature point information of the main building and auxiliary structures with different bottom elevations and main body heights, the feature point information of windows belonging to different floors, and the feature point information of slopes that do not belong to the same slope are extracted by type and stored in feature point files according to rules. Among them, the feature points of the main building belonging to the same block are saved in a separate feature point file, the feature points of the auxiliary structures belonging to the same block are saved in a separate feature point file, the feature points of windows with the same window sill elevation and window height on the same floor are saved in a separate feature point file, and the feature points of the same slope are saved in a separate feature point file.

[0118] like Figure 3 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for feature point classification, extraction, numbering, and annotation:

[0119] Exporting the feature point extraction model: Define sample input functions based on feature points of the main building, auxiliary structures, sunlight windows, and rooftop, which are used to train and evaluate the feature point extraction model; define hook callback functions and register them in the estimator framework to output detailed information during the training process; establish a for loop and perform training and evaluation operations on the feature point extraction model inside the loop, while outputting relevant information; after training is completed, export the feature point extraction model.

[0120] Feature point extraction, numbering, and remarks: Based on the feature point extraction model, feature points constituting the main building structure, auxiliary structures, windows, and rooftops are automatically extracted through calculation. The automatically extracted feature points are checked, and erroneous feature points are corrected through human-computer interaction. For feature points belonging to the same block constituting the main building structure or auxiliary structures, window features on the same floor, and feature points constituting the same rooftop, a starting point is randomly determined, and then the feature points are numbered sequentially. Remarks are added as needed; otherwise, they are not required. The formula for automatically extracting feature points constituting the main building structure, auxiliary structures, windows, and rooftops is:

[0121]

[0122] In the formula:

[0123] U(s,a) Feature point extraction probability

[0124] c puct exploration level constant

[0125] P(s,a) Prior probability

[0126] N(s,b) is the number of times the parent node has been visited.

[0127] N(s,a) represents the number of visits to a certain child node.

[0128] When extracting feature point information that constitutes the main building, auxiliary structures, and sunlight windows, at least one feature point at the bottom edge and one feature point at the top edge should be ensured.

[0129] like Figure 4 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for automatically constructing the solar radiation analysis model:

[0130] Automatically construct building main structure, auxiliary structure, and window models: Retrieve all feature point files of the building main structure, auxiliary structure, and windows stored in a specified path and named according to a specific naming method, and read each feature point file one by one; traverse the feature point data of each feature point file line by line, and store the feature point data of each feature point file into different arrays; calculate the bottom elevation and main body height of each building main structure and auxiliary structure, and the window sill elevation and window height of each floor / each window based on the z-coordinate values ​​of the feature points in different arrays; connect the feature points in sequence using contour lines based on the (x, y) coordinate values ​​of the feature points belonging to the same block extracted from the array, generate the outer contour line of the building main structure or the outer contour line of the auxiliary structure, and establish the building main structure and auxiliary structure models based on the bottom elevation and main body height of the contour lines; extract the (x, y) coordinate values ​​of the left and right window sill feature points of each window on the same floor from the array, and establish the window model using a two-point window insertion method based on the window sill feature point coordinate values, window sill elevation, and window height;

[0131] Automatically construct the slope top model: retrieve all feature point files of the slope top stored in the specified path and named according to a specific naming method, and read each feature point file one by one; then traverse the feature point data in the feature point file line by line, and store the feature point data of each slope top feature point file into different arrays; generate a three-dimensional mesh surface by connecting the slope top feature points in sequence to construct the slope top model.

[0132] like Figure 5 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for automatically constructing the main building, auxiliary structures, and solar windows:

[0133] Retrieve feature point files for the main building, ancillary structures, and sun windows: Retrieve and obtain all feature point files for the main building, ancillary structures, and sun windows, and iterate through and read the feature point files for the main building, ancillary structures, and sun windows one by one;

[0134] Read feature point coordinates and store them in an array: Iterate through the feature point data in each feature point file, obtain the number and coordinate information (x, y, z) of each feature point in each feature point file, and determine whether coordinate unit conversion is required. If it is determined that coordinate unit conversion is required, after iterating through and obtaining the feature point coordinate information, convert the coordinate units according to the formula. After conversion, store the number and coordinate information of feature points belonging to the main body and auxiliary structures of the same block, and feature points belonging to each sun window on the same floor, in the corresponding arrays in order. One feature point file corresponds to one array.

[0135] The formula for converting coordinate units is:

[0136] x = n × x0

[0137] y = n × y0

[0138] z = n × z0

[0139] In the formula:

[0140] x0, y0, and z0 represent the coordinate values ​​of feature points extracted from the point cloud;

[0141] n represents the scaling factor;

[0142] x, y, and z represent the coordinate values ​​corresponding to the plotted solar radiation model;

[0143] When the original coordinate values ​​of the point cloud data are in meters, and the coordinate values ​​used to draw the solar radiation model are in millimeters, then n is set to 1000 for coordinate unit conversion.

[0144] When it is determined that no coordinate unit conversion is required, the feature point number and coordinate value information (x, y, z) of the feature point file are obtained, and the feature points belonging to the same block of the main building and auxiliary structure, and the feature points belonging to each sun window on the same floor are directly stored in the same array in order.

[0145] Calculate the base elevation: Extract the z-coordinate values ​​of feature points belonging to the same building block and its auxiliary structures, as well as feature points of windows with the same sill elevation and window height, from the array, and add them to their corresponding sequences; traverse the sequence of each feature point belonging to the same building block and its auxiliary structures, as well as each feature point of windows with the same sill elevation and window height, to calculate the base elevation of each different building block and its auxiliary structures, and the sill elevation of the windows; the specific method is as follows:

[0146] Iterate through the feature points of the main building and auxiliary structures belonging to the same block, as well as the feature points of windows with the same sill elevation and window height, and add them to their respective sequences. Sort the z-coordinate values ​​of the feature points in the sequences, set a threshold Δh, with the threshold range between 0m and 1m, and calculate the average z-coordinate of the bottom feature points of all the bottom feature points of the main building and auxiliary structures belonging to the same block, as well as the bottom feature points of windows with the same sill elevation and window height within the threshold range. Calculate the bottom elevation h of the windows of different blocks, the main building and auxiliary structures, and different floors. 底标高 The specific calculation process is as follows:

[0147] First, determine the coordinates of the lowest point:

[0148] z min =min(z) i (i = 1, 2, ..., n)

[0149] Secondly, based on the set threshold Δh, feature points participating in the calculation are selected. If the j-th feature point z j If the difference between the coordinates of the lowest point and the coordinates of the lowest point is within the threshold range, then it participates in the next step h. 底标高 The calculation determines whether the feature points participate in h. 底标高 The formula for the calculation is:

[0150] z j -z min ≤Δh(j=1,2,…,m)

[0151] Finally, calculate the bottom elevation h. 底标高 :

[0152]

[0153] When the threshold Δh is 0, then h 底标高 The calculation formula is:

[0154] h 底标高 =min(z) i (i = 1, 2, ..., n)

[0155] In the formula:

[0156] h 底标高 This refers to the bottom elevation of the main building structure or its ancillary structures, or the windowsill elevation of a window that receives sunlight.

[0157] z i Let z be the z-coordinate of the i-th point, min be the minimum value, and n be the total number of feature points;

[0158] z j This represents the z-coordinate value of the j-th bottom feature point that is within the threshold range and participates in the calculation; m represents the number of bottom feature points participating in the calculation.

[0159] Calculate the main building height and window height: Extract the z-coordinate values ​​of feature points belonging to the same building block and its auxiliary structures, as well as feature points of windows with the same sill elevation and window height, from the array, and add them to their corresponding sequences; traverse the sequence of each feature point belonging to the same building block and its auxiliary structures, as well as each feature point of windows with the same sill elevation and window height, and calculate the main building height and window height of each different building block and its auxiliary structures; the specific method is as follows:

[0160] Iterate through the feature points of the main building and auxiliary structures belonging to the same block, as well as the feature points of windows with the same sill elevation and window height, using their z-coordinates. Add these z-coordinates to their respective sequences, sort the z-coordinate values ​​of the feature points in the sequences, and set a threshold Δh, with the threshold range between 0m and 1m. Calculate the average z-coordinates of the top feature points of all the top feature points of the main building and auxiliary structures belonging to the same block, as well as the top feature points of windows with the same sill elevation and window height within the threshold range. Calculate the top height h of the windows for different blocks of the main building, auxiliary structures, and different floors. 顶部高 Determine the main body height and window height h. 主体高 for h 顶部高 with h 底标高 The difference is calculated as follows:

[0161] First, determine the coordinates of the highest point:

[0162] z max =max(z) i (i = 1, 2, ..., n)

[0163] Secondly, based on the set threshold Δh, feature points participating in the calculation are selected. If the j-th feature point z j If the difference between the coordinates of the highest point and the coordinates of the highest point is within the threshold range, then it participates in the next step h. 顶部高 The calculation determines whether the feature points participate in h. 顶部高 The formula for the calculation is:

[0164] z max -z j ≤Δh(i=1,2,…,m)

[0165] Then, calculate the top height h. 顶部高 :

[0166]

[0167] When the threshold Δh is 0, then h 顶部高 The calculation formula is:

[0168] h 顶部高 =max(z)i (i = 1, 2, ..., n)

[0169] Finally, calculate the main body height h. 主体高 :

[0170] h 主体高 =h 顶部高 -h 底标高

[0171] In the formula:

[0172] h 主体高 The height of the main building and its ancillary structures, and the height of the windows for sunlight; h 顶部高 h is the height of the top of the main building, auxiliary structures, and windows. 底标高 This refers to the bottom elevation of the main building structure or its ancillary structures, or the windowsill elevation of a window that receives sunlight.

[0173] z i Let z be the z-coordinate of the i-th feature point, max be the maximum value, and n be the total number of feature points;

[0174] z j This represents the z-coordinate value of the j-th top feature point that is within the threshold range and participates in the calculation; m represents the number of top feature points participating in the calculation.

[0175] Determine the modeling type: Determine the model type according to the main building, auxiliary structures, and sunlight windows, and construct the model based on the determined model type;

[0176] Building Main Structure and Ancillary Structure Model Construction: Based on the feature point coordinate information extracted from the array, the bottom elevation (h) of the main structure and ancillary structures is obtained from the bottom elevation calculation step. 底标高 The numerical values ​​for calculating the main building height and window height are obtained from the calculation steps. 主体高 The system uses numerical values ​​to draw building outlines and automatically constructs building models corresponding to feature point files of each building and its ancillary structures.

[0177] Sunlight window model construction: Based on the coordinate values ​​of feature points of each sunlight window on the same floor extracted from the array, the window sill elevation (h) obtained in the bottom elevation step is calculated. 底标高 The window height (h) is obtained by calculating the main body height and window height. 主体高 The numerical values ​​are used to extract the windowsill points of the sunshine windows, and one or more sunshine window models corresponding to each sunshine window feature point file are automatically constructed.

[0178] like Figure 6 As shown, the modeling method for a high-precision solar radiation analysis model based on point clouds includes the following steps in constructing the building main body and auxiliary structure models:

[0179] Drawing building outlines: Extract the (x, y) coordinates of each feature point belonging to the same block of the main building and its auxiliary structures from the array; automatically connect the feature points with outlines according to the coordinate order to generate the outer outline of the main building or the auxiliary structures; determine whether the outline is closed according to requirements; when it is determined that the outline needs to be closed, connect the last feature point to the first feature point end to end and set it as a closed outline; when it is determined that the outline does not need to be closed, the last feature point does not need to be connected to the first feature point.

[0180] Model construction: Set the bottom elevation value of the contour line to the bottom elevation (h) obtained from the bottom elevation calculation step. 底标高 The thickness value of the outline is set to the main body height (h) obtained from the steps of calculating the main body height and window height. 主体高 Set the bottom elevation value of the outline to the bottom elevation (h) obtained from the bottom elevation calculation step. 底标高 After that, the block is modeled by stretching it up according to the outline. The height of the stretched block is the main body height (h) obtained from the steps of calculating the main body height and window height. 主体高 );

[0181] When the outer contour shape of balconies on each floor is consistent and the spacing between balconies on different floors is within the tolerance range, the following method is adopted: first, a balcony model of a certain floor is established based on the contour line, bottom elevation, and main structure height; then, based on the feature point file information, the spacing value △H of the balconies in the Z direction is calculated, and the balconies of each floor are copied according to the spacing value array.

[0182] The calculation formula is:

[0183]

[0184] In the formula:

[0185] z m z m+1 This indicates the bottom elevation or top elevation of the balcony on the m-th or m+1-th floor;

[0186] n represents the number of balcony floors involved in the calculation and modeling;

[0187] △H represents the average spacing value.

[0188] like Figure 7 As shown, a high-precision solar radiation analysis modeling method based on point clouds is described, wherein the construction of the solar radiation window model includes the following steps:

[0189] Extracting left and right window sill points of sunlit windows: Extract the (x,y) coordinate values ​​of feature points of each sunlit window on the same floor from the array to distinguish between left and right window sill points;

[0190] Insert windows: Using the left and right endpoints of the window as the start and end points, insert windows at these two points to provide sunlight. Set the windowsill elevation to the windowsill elevation (h) obtained in the calculation of the bottom elevation step. 底标高 Set the window height to the height of the main body and the window height obtained in the window height calculation step (h). 主体高 );

[0191] like Figure 8 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for automatically constructing the hilltop model:

[0192] Retrieve slope crest feature point files: Based on the file name and storage path of the feature point files, retrieve all slope crest feature point files, and then iterate through and read the slope crest feature point files one by one;

[0193] Read feature point coordinates and store them in an array: Iterate through the feature point data in each feature point file line by line, obtain the coordinates (x, y, z) and remarks of each feature point in each feature point file, and store the coordinates and remarks of the feature points that constitute the top of the slope in the corresponding array in order; Perform coordinate unit conversion according to the steps of reading feature point coordinates and storing them in an array in the automatic construction of building main body, auxiliary structure and sunlight window model;

[0194] To create a 3D mesh surface, connect all feature points on the same slope point by point. The specific method is as follows: traverse the feature points in the array and extract them, connect them to construct a 3D surface, and calculate the slope and aspect of the 3D surface; calculate and obtain 3D surfaces with the same slope and aspect, extract the points that make up the 3D surface, and store their (x, y, z) values ​​in an array; iterate through the calculations, storing points on the same slope surface in the same array, resulting in a separate array for each slope surface; remove duplicate coordinates of the feature points (x, y, z) in the slope coordinate array; calculate the order of the coordinate points in each array, and connect the feature points on the slope surface with the same slope point one by one according to the order to generate a 3D mesh surface.

[0195] In specific implementation cases, the ancillary structures include: balconies, eaves, columns, air conditioning partitions, unit partitions, moldings, sunshades, parapet walls of the roof, stairwells leading to the roof, water tank rooms, decorative components, and elevator machine rooms.

[0196] In specific implementation cases: the feature point information is stored in the following ways: feature point information of building main bodies and auxiliary structures with different bottom elevations and main body heights, feature point information of sun windows belonging to different floors, and feature point information of slopes not belonging to the same slope are saved to different feature point files. The feature point files are named according to the feature point type and are distinguished by adding suffixes, etc. The feature point information of building main bodies and auxiliary structures with different bottom elevations and main body heights, feature point information of sun windows belonging to different floors, and feature point information of slopes not belonging to the same slope are stored in the same file path according to category, and the folder is named according to the feature point type.

[0197] In specific implementation cases: the method of establishing a model based on the bottom elevation of the outline and the height of the main body includes: modeling by setting the bottom elevation of the outline and the height of the main body; or modeling by setting the bottom elevation of the outline and then raising the block, where the height of the raised block is the height of the main body.

[0198] Example 2:

[0199] Figure 1 A modeling method for high-precision solar radiation analysis based on point clouds, such as Figure 1 As shown, the method includes the following steps:

[0200] Extraction and storage of feature point information: A 3D laser scanner is used for field data acquisition to obtain building point cloud data; the scanned point cloud is processed by point cloud data registration, coordinate system transformation, noise reduction and thinning; feature point information of buildings / structures is extracted according to rules and categories, the feature point number and coordinate information (x, y, z) are exported, and feature point files of the same type are named according to the same rules or stored in the same file path;

[0201] Automatically construct a solar radiation analysis model: Traverse the feature point files, and based on the feature point type, feature point number, coordinate value, and remarks, automatically construct the building main body, auxiliary structure model, solar window model, and roof model corresponding to each feature point file. Each individual model is placed in the corresponding three-dimensional space according to the coordinate value, and automatically combined to generate a complete high-precision solar radiation analysis model.

[0202] like Figure 2 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for extracting and storing feature point information:

[0203] Feature point classification, extraction, numbering, and annotation: Extract feature points from the point cloud data for the main building, auxiliary structures, sunlight windows, and rooftops, and collect feature point coordinate information;

[0204] Feature point information storage: The feature point information of the main building and auxiliary structures with different bottom elevations and main body heights, the feature point information of windows belonging to different floors, and the feature point information of slopes that do not belong to the same slope are extracted by type and stored in feature point files according to rules. Among them, the feature points of the main building belonging to the same block are saved in a separate feature point file, the feature points of the auxiliary structures belonging to the same block are saved in a separate feature point file, the feature points of windows with the same window sill elevation and window height on the same floor are saved in a separate feature point file, and the feature points of the same slope are saved in a separate feature point file.

[0205] like Figure 3 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for feature point classification, extraction, numbering, and annotation:

[0206] Exporting the feature point extraction model: Define sample input functions based on feature points of the main building, auxiliary structures, sunlight windows, and rooftop, which are used to train and evaluate the feature point extraction model; define hook callback functions and register them in the estimator framework to output detailed information during the training process; establish a for loop and perform training and evaluation operations on the feature point extraction model inside the loop, while outputting relevant information; after training is completed, export the feature point extraction model.

[0207] Feature point extraction, numbering, and remarks: Based on the feature point extraction model, feature points constituting the main building structure, auxiliary structures, windows, and rooftops are automatically extracted through calculation. The automatically extracted feature points are checked, and erroneous feature points are corrected through human-computer interaction. For feature points belonging to the same block constituting the main building structure or auxiliary structures, window features on the same floor, and feature points constituting the same rooftop, a starting point is randomly determined, and then the feature points are numbered sequentially. Remarks are added as needed; otherwise, they are not required. The formula for automatically extracting feature points constituting the main building structure, auxiliary structures, windows, and rooftops is:

[0208]

[0209] In the formula:

[0210] U(s,a) Feature point extraction probability

[0211] c puct exploration level constant

[0212] P(s,a) Prior probability

[0213] N(s,b) is the number of times the parent node has been visited.

[0214] N(s,a) represents the number of visits to a certain child node.

[0215] When extracting feature point information that constitutes the main building, auxiliary structures, and sunlight windows, at least one feature point at the bottom edge and one feature point at the top edge should be ensured.

[0216] The feature point numbers, x, y, and z coordinates stored in the feature point file corresponding to the main building outline are shown in the table below:

[0217] Number X coordinate Y coordinate Z coordinate Remark ZT001 99213.711 286497.760 13.126 ZT002 99259.227 286497.451 2.770 ZT003 99259.176 286484.909 2.772 ZT004 99213.637 286485.304 20.784

[0218] The feature point numbers, x, y, and z coordinates stored in the feature point file corresponding to the outline of the building's attached structures (balconies) are shown in the table below:

[0219]

[0220]

[0221]

[0222] The feature point numbers, x, y, and z coordinates stored in the feature point file corresponding to the outline of the building's auxiliary structures (eaves) are shown in the table below:

[0223]

[0224] like Figure 4 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for automatically constructing the solar radiation analysis model:

[0225] Automatically construct building main structure, auxiliary structure, and window models: Retrieve all feature point files of the building main structure, auxiliary structure, and windows stored in a specified path and named according to a specific naming method, and read each feature point file one by one; traverse the feature point data of each feature point file line by line, and store the feature point data of each feature point file into different arrays; calculate the bottom elevation and main body height of each building main structure and auxiliary structure, and the window sill elevation and window height of each floor / each window based on the z-coordinate values ​​of the feature points in different arrays; connect the feature points in sequence using contour lines based on the (x, y) coordinate values ​​of the feature points belonging to the same block extracted from the array, generate the outer contour line of the building main structure or the outer contour line of the auxiliary structure, and establish the building main structure and auxiliary structure models based on the bottom elevation and main body height of the contour lines; extract the (x, y) coordinate values ​​of the left and right window sill feature points of each window on the same floor from the array, and establish the window model using a two-point window insertion method based on the window sill feature point coordinate values, window sill elevation, and window height;

[0226] Automatically construct the slope top model: retrieve all feature point files of the slope top stored in the specified path and named according to a specific naming method, and read each feature point file one by one; then traverse the feature point data in the feature point file line by line, and store the feature point data of each slope top feature point file into different arrays; generate a three-dimensional mesh surface by connecting the slope top feature points in sequence to construct the slope top model.

[0227] like Figure 5 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for automatically constructing the main building, auxiliary structures, and solar windows:

[0228] Retrieve feature point files for the main building, ancillary structures, and sun windows: Retrieve and obtain all feature point files for the main building, ancillary structures, and sun windows, and iterate through and read the feature point files for the main building, ancillary structures, and sun windows one by one;

[0229] Read feature point coordinates and store them in an array: Iterate through the feature point data in each feature point file, obtain the number and coordinate information (x, y, z) of each feature point in each feature point file, and determine whether coordinate unit conversion is required. If it is determined that coordinate unit conversion is required, after iterating through and obtaining the feature point coordinate information, convert the coordinate units according to the formula. After conversion, store the number and coordinate information of feature points belonging to the main body and auxiliary structures of the same block, and feature points belonging to each sun window on the same floor, in the corresponding arrays in order. One feature point file corresponds to one array.

[0230] The coordinates of the main building are stored in an array in numerical order as follows:

[0231] numpy.array([

[0232] [99213.711,286497.760,13.126],

[0233] [99259.227,286497.451,2.770],

[0234] [99259.176,286484.909,2.772],

[0235] [99213.637,286485.304,20.784] ])

[0237] The coordinates of the building's ancillary structures (balconies) are stored in an array in numerical order as follows:

[0238] numpy.array([

[0239] [99217.551,286497.750,20.764],

[0240] [99217.529,286498.953,2.956],

[0241] [99220.891,286498.936,17.839],

[0242] [99220.898,286497.718,15.328],

[0243] [99223.246,286497.702,15.280],

[0244] [99223.267,286498.913,14.909],

[0245] [99225.714,286498.897,14.858],

[0246] [99225.708,286497.682,20.780],

[0247] [99232.842,286497.632,20.790],

[0248] [99232.842,286498.855,14.779],

[0249] [99235.307,286498.824,11.919],

[0250] [99235.306,286497.618,12.297],

[0251] [99237.637,286497.601,11.751],

[0252] [99237.642,286498.794,9.170],

[0253] [99240.104,286498.788,9.402],

[0254] [99240.127,286497.580,20.763],

[0255] [99247.243,286497.541,20.803],

[0256] [99247.242,286498.739,9.380],

[0257] [99249.709,286498.729,9.266],

[0258] [99249.703,286497.534,9.072],

[0259] [99252.002,286497.502,6.769],

[0260] [99252.031,286498.720,2.965],

[0261] [99255.413,286498.693,3.126],

[0262] [99255.388,286497.482,20.767],

[0263] [99259.216,286492.418,20.796],

[0264] [99260.423,286492.399,6.550],

[0265] [99260.410,286489.943,2.975],

[0266] [99259.195,286489.943,20.785],

[0267] [99254.429,286484.955,20.790],

[0268] [99254.401,286483.721,11.946],

[0269] [99247.141,286483.781,11.858],

[0270] [99247.160,286484.990,20.808],

[0271] [99240.010,286485.045,12.354],

[0272] [99239.995,286483.816,14.720],

[0273] [99232.732,286483.876,14.926],

[0274] [99232.753,286485.090,20.782],

[0275] [99225.631,286485.138,20.814],

[0276] [99225.618,286483.922,9.064],

[0277] [99218.346,286483.976,2.961],

[0278] [99218.362,286485.197,14.966],

[0279] [99213.647,286490.247,20.247],

[0280] [99212.472,286490.264,2.944],

[0281] [99212.470,286492.718,6.543],

[0282] [99213.676,286492.738,20.776] ])

[0284] The coordinate values ​​of the building's auxiliary structures (eaves) are stored in an array in numerical order as follows:

[0285] numpy.array([

[0286] [99213.020,286498.425,20.986],

[0287] [99216.854,286498.430,20.771],

[0288] [99216.914,286499.641,20.823],

[0289] [99226.425,286499.583,20.927],

[0290] [99226.348,286498.355,20.775],

[0291] [99232.190,286498.358,20.965],

[0292] [99232.207,286499.549,20.821],

[0293] [99240.810,286499.481,20.955],

[0294] [99240.798,286498.239,20.778],

[0295] [99246.507,286498.188,20.809],

[0296] [99246.548,286499.425,20.887],

[0297] [99256.092,286499.333,21.003],

[0298] [99256.029,286498.170,20.778],

[0299] [99259.988,286498.093,20.996],

[0300] [99259.922,286493.124,20.996],

[0301] [99261.151,286493.065,20.962],

[0302] [99261.043,286489.287,20.994],

[0303] [99259.889,286489.309,20.965],

[0304] [99259.861,286484.162,20.984],

[0305] [99255.116,286484.239,20.985],

[0306] [99255.057,286482.994,20.972],

[0307] [99246.448,286483.117,20.948],

[0308] [99246.446,286484.329,20.796],

[0309] [99240.675,286484.358,21.000],

[0310] [99240.564,286483.095,20.985],

[0311] [99232.011,286483.220,20.983],

[0312] [99232.009,286484.410,21.012],

[0313] [99226.300,286484.459,21.031],

[0314] [99226.287,286483.180,20.953],

[0315] [99217.641,286483.253,20.982],

[0316] [99217.629,286484.471,20.995],

[0317] [99212.908,286484.502,20.984],

[0318] [99212.971,286489.612,20.833],

[0319] [99211.759,286489.615,20.970],

[0320] [99211.761,286493.409,20.957],

[0321] [99212.999,286493.460,20.936] ])

[0323] The formula for converting coordinate units is:

[0324] x = n × x0

[0325] y = n × y0

[0326] z = n × z0

[0327] In the formula:

[0328] x0, y0, and z0 represent the coordinate values ​​of feature points extracted from the point cloud;

[0329] n represents the scaling factor;

[0330] x, y, and z represent the coordinate values ​​corresponding to the plotted solar radiation model;

[0331] When the original coordinate values ​​of the point cloud data are in meters, and the coordinate values ​​used to draw the solar radiation model are in millimeters, then n is set to 1000 for coordinate unit conversion.

[0332] The following is the array after converting the coordinate units of the main building from meters to millimeters:

[0333] numpy.array([

[0334] [99213711,286497760,13126],

[0335] [99259227,286497451,2770],

[0336] [99259176,286484909,2772],

[0337] [99213637,286485304,20784] ])

[0339] The following array, after converting the coordinate units of the building's auxiliary structures (balconies) from meters to millimeters, is as follows:

[0340] numpy.array([

[0341] [99217551,286497750,20764],

[0342] [99217529,286498953,2956],

[0343] [99220891,286498936,17839],

[0344] [99220898,286497718,15328],

[0345] [99223246,286497702,15280],

[0346] [99223267,286498913,14909],

[0347] [99225714,286498897,14858],

[0348] [99225708,286497682,20780],

[0349] [99232842,286497632,20790],

[0350] [99232842,286498855,14779],

[0351] [99235307,286498824,11919],

[0352] [99235306,286497618,12297],

[0353] [99237637,286497601,11751],

[0354] [99237642,286498794,9170],

[0355] [99240104,286498788,9402],

[0356] [99240127,286497580,20763],

[0357] [99247243,286497541,20803],

[0358] [99247242,286498739,9380],

[0359] [99249709,286498729,9266],

[0360] [99249703,286497534,9072],

[0361] [99252002,286497502,6769],

[0362] [99252031,286498720,2965],

[0363] [99255413,286498693,3126],

[0364] [99255388,286497482,20767],

[0365] [99259216,286492418,20796],

[0366] [99260423,286492399,6550],

[0367] [99260410,286489943,2975],

[0368] [99259195,286489943,20785],

[0369] [99254429,286484955,20790],

[0370] [99254401,286483721,11946],

[0371] [99247141,286483781,11858],

[0372] [99247160,286484990,20808],

[0373] [99240010,286485045,12354],

[0374] [99239995,286483816,14720],

[0375] [99232732,286483876,14926],

[0376] [99232753,286485090,20782],

[0377] [99225631,286485138,20814],

[0378] [99225618,286483922,9064],

[0379] [99218346,286483976,2961],

[0380] [99218362,286485197,14966],

[0381] [99213647,286490247,20247],

[0382] [99212472,286490264,2944],

[0383] [99212470,286492718,6543],

[0384] [99213676,286492738,20776] ])

[0386] The following array, after converting the coordinate units of the building's auxiliary structures (eaves) from meters to millimeters, is as follows:

[0387] numpy.array([

[0388] [99213020,286498425,20986],

[0389] [99216854,286498430,20771],

[0390] [99216914,286499641,20823],

[0391] [99226425,286499583,20927],

[0392] [99226348,286498355,20775],

[0393] [99232190,286498358,20965],

[0394] [99232207,286499549,20821],

[0395] [99240810,286499481,20955],

[0396] [99240798,286498239,20778],

[0397] [99246507,286498188,20809],

[0398] [99246548,286499425,20887],

[0399] [99256092,286499333,21003],

[0400] [99256029,286498170,20778],

[0401] [99259988,286498093,20996],

[0402] [99259922,286493124,20996],

[0403] [99261151,286493065,20962],

[0404] [99261043,286489287,20994],

[0405] [99259889,286489309,20965],

[0406] [99259861,286484162,20984],

[0407] [99255116,286484239,20985],

[0408] [99255057,286482994,20972],

[0409] [99246448,286483117,20948],

[0410] [99246446,286484329,20796],

[0411] [99240675,286484358,21000],

[0412] [99240564,286483095,20985],

[0413] [99232011,286483220,20983],

[0414] [99232009,286484410,21012],

[0415] [99226300,286484459,21031],

[0416] [99226287,286483180,20953],

[0417] [99217641,286483253,20982],

[0418] [99217629,286484471,20995],

[0419] [99212908,286484502,20984],

[0420] [99212971,286489612,20833],

[0421] [99211759,286489615,20970],

[0422] [99211761,286493409,20957],

[0423] [99212999,286493460,20936] ])

[0425] When it is determined that no coordinate unit conversion is required, the feature point number and coordinate value information (x, y, z) of the feature point file are obtained, and the feature points belonging to the same block of the main building and auxiliary structure, and the feature points belonging to each sun window on the same floor are directly stored in the same array in order.

[0426] Calculate the base elevation: Extract the z-coordinate values ​​of feature points belonging to the same building block and its auxiliary structures, as well as feature points of windows with the same sill elevation and window height, from the array, and add them to their corresponding sequences; traverse the sequence of each feature point belonging to the same building block and its auxiliary structures, as well as each feature point of windows with the same sill elevation and window height, to calculate the base elevation of each different building block and its auxiliary structures, and the sill elevation of the windows; the specific method is as follows:

[0427] Iterate through the feature points of the main building and auxiliary structures belonging to the same block, as well as the feature points of windows with the same sill elevation and window height, and add them to their respective sequences. Sort the z-coordinate values ​​of the feature points in the sequences, set a threshold Δh, with the threshold range between 0m and 1m, and calculate the average z-coordinate of the bottom feature points of all the bottom feature points of the main building and auxiliary structures belonging to the same block, as well as the bottom feature points of windows with the same sill elevation and window height within the threshold range. Calculate the bottom elevation h of the windows of different blocks, the main building and auxiliary structures, and different floors. 底标高 The specific calculation process is as follows:

[0428] First, determine the coordinates of the lowest point:

[0429] z min =min(z) i (i = 1, 2, ..., n)

[0430] In this embodiment, the feature point of the lowest point of the main building is calculated and determined to be numbered ZT002, and its corresponding z-coordinate is z. min It is 2770mm;

[0431] In this embodiment, the feature point of the lowest point of the building's auxiliary structure (balcony) is numbered yt042, and its corresponding z-coordinate is z. min The value is 2944mm.

[0432] In this embodiment, the feature point of the lowest point of the building's auxiliary structure (eaves) is numbered FY002, and its corresponding z-coordinate is z. min The value is 20771 mm.

[0433] Secondly, based on the set threshold Δh, feature points participating in the calculation are selected. If the j-th feature point z j If the difference between the coordinates of the lowest point and the coordinates of the lowest point is within the threshold range, then it participates in the next step h. 底标高 The calculation determines whether the feature points participate in h. 底标高 The formula for the calculation is:

[0434] z j -z min ≤Δh(j=1,2,…,m)

[0435] Finally, calculate the bottom elevation h. 底标高 :

[0436]

[0437] With Δh = 50mm, the feature points of the main building involved in the bottom elevation calculation in this example are numbered ZT002 and ZT003 respectively, with corresponding z-coordinates of 2770mm and 2771mm respectively. The calculated bottom elevation values ​​are as follows:

[0438] h 底标高 = (2770 + 2772) ÷ 2 = 2771

[0439] With Δh = 50mm, the feature points of the building attachment structure (balcony) involved in the bottom elevation calculation in this example are numbered yt042, yt002, yt039, yt022, and yt027, respectively, with corresponding z-coordinates of 2944mm, 2956mm, 2961mm, 2965mm, and 2975mm. The calculated bottom elevation values ​​are as follows:

[0440] h 底标高 =(2944+2956+2961+2965+2975)÷5=2960mm

[0441] With Δh = 50mm, the feature points of the building attachments (eaves) involved in the bottom elevation calculation in this example are numbered FY002, FY005, FY009, FY013, FY023, FY010, and FY007, respectively, with corresponding z-coordinates of 20771mm, 20775mm, 20778mm, 20778mm, 20796mm, 20809mm, and 20821mm. The calculated bottom elevation values ​​are as follows:

[0442] h 底标高 =(20771+20775+20778+20778+20796+20809+20821)÷7=20790mm

[0443] When the threshold Δh is 0, then h 底标高 The calculation formula is:

[0444] h 底标高 =min(z) i (i = 1, 2, ..., n)

[0445] In the formula:

[0446] h 底标高 This refers to the bottom elevation of the main building structure or its ancillary structures, or the windowsill elevation of a window that receives sunlight.

[0447] z i Let z be the z-coordinate of the i-th point, min be the minimum value, and n be the total number of feature points;

[0448] z j This represents the z-coordinate value of the j-th bottom feature point that is within the threshold range and participates in the calculation; m represents the number of bottom feature points participating in the calculation.

[0449] Calculate the main building height and window height: Extract the z-coordinate values ​​of feature points belonging to the same building block and its auxiliary structures, as well as feature points of windows with the same sill elevation and window height, from the array, and add them to their corresponding sequences; traverse the sequence of each feature point belonging to the same building block and its auxiliary structures, as well as each feature point of windows with the same sill elevation and window height, and calculate the main building height and window height of each different building block and its auxiliary structures; the specific method is as follows:

[0450] Iterate through the feature points of the main building and auxiliary structures belonging to the same block, as well as the feature points of windows with the same sill elevation and window height, using their z-coordinates. Add these z-coordinates to their respective sequences, sort the z-coordinate values ​​of the feature points in the sequences, and set a threshold Δh, with the threshold range between 0m and 1m. Calculate the average z-coordinates of the top feature points of all the top feature points of the main building and auxiliary structures belonging to the same block, as well as the top feature points of windows with the same sill elevation and window height within the threshold range. Calculate the top height h of the windows for different blocks of the main building, auxiliary structures, and different floors. 顶部高 Determine the main body height and window height h. 主体高 for h 顶部高 with h 底标高 The difference is calculated as follows:

[0451] First, determine the coordinates of the highest point:

[0452] z max =max(z) i (i = 1, 2, ..., n)

[0453] In this embodiment, the feature point of the highest point of the main building is numbered ZT004, and its corresponding z-coordinate is z. max The value is 20784 mm.

[0454] In this embodiment, the feature point of the highest point of the building's auxiliary structure (balcony) is numbered yt037, and its corresponding z-coordinate is z. max The value is 20814 mm.

[0455] In this embodiment, the feature point of the highest point of the building's auxiliary structure (eaves) is numbered FY028, and its corresponding z-coordinate is z. max The value is 21031 mm.

[0456] Secondly, based on the set threshold Δh, feature points participating in the calculation are selected. If the j-th feature point z j If the difference between the coordinates of the highest point and the coordinates of the highest point is within the threshold range, then it participates in the next step h. 顶部高 The calculation determines whether the feature points participate in h. 顶部高 The formula for the calculation is:

[0457] z max -z j ≤Δh(i=1,2,…,m)

[0458] Then, calculate the top height h. 顶部高 :

[0459]

[0460] With Δh = 50mm, the feature point of the building main body involved in the top height calculation in this example is numbered ZT004, and the corresponding z-coordinate is 20784mm. The calculated value of the top height of the building main body is as follows:

[0461]

[0462] With Δh = 50mm, the feature points of the building attachment structure (balcony) involved in the top height calculation in this example are numbered yt001, yt024, yt044, yt008, yt036, yt028, yt029, yt009, yt025, yt017, yt032, and yt037, respectively, with corresponding z-coordinates of 20764mm, 20767mm, 20776mm, 20780mm, 20782mm, 20785mm, 20790mm, 20790mm, 20796mm, 20803mm, 20808mm, and 20814mm. The calculated top height value of the building attachment structure (balcony) is as follows:

[0463]

[0464] With Δh = 50mm, the feature points of the building attachments (eaves) involved in the calculation of the top height in this example are numbered FY030, FY026, FY019, FY032, FY020, FY025, FY001, FY017, FY031, FY014, FY015, FY024, FY012, FY027, and FY028, respectively. Their corresponding z-coordinates are 20982mm, 20983mm, 20984mm, 20984mm, 20985mm, 20985mm, 20986mm, 20994mm, 20995mm, 20996mm, 20996mm, 21.000mm, 21003mm, 21012mm, and 21031. The calculated values ​​of the top height of the building attachments (eaves) are as follows:

[0465]

[0466] When the threshold Δh is 0, then h 顶部高 The calculation formula is:

[0467] h 顶部高 =max(z) i (i = 1, 2, ..., n)

[0468] Finally, calculate the main body height h. 主体高 :

[0469] h 主体高 =h 顶部高 -h 底标高

[0470] In this embodiment, the main height of the building is:

[0471] h 主体高 =h 顶部高 -h 底标高 =20784-2771=18013mm

[0472] In this embodiment, the main height of the building's auxiliary structure (balcony) is:

[0473] h 主体高 =h 顶部高 -h 底标高 =20788-2960=17.828mm

[0474] In this embodiment, the main height of the building's auxiliary structure (eaves) is:

[0475] h 主体高 =h 顶部高 -h 底标高 =20994-20790=204mm

[0476] In the formula:

[0477] h 主体高 The height of the main building and its ancillary structures, and the height of the windows for sunlight; h 顶部高 h is the height of the top of the main building, auxiliary structures, and windows. 底标高 This refers to the bottom elevation of the main building structure or its ancillary structures, or the windowsill elevation of a window that receives sunlight.

[0478] z i Let z be the z-coordinate of the i-th feature point, max be the maximum value, and n be the total number of feature points;

[0479] z j This represents the z-coordinate value of the j-th top feature point that is within the threshold range and participates in the calculation; m represents the number of top feature points participating in the calculation.

[0480] Determine the modeling type: Determine the model type according to the main building, auxiliary structures, and sunlight windows, and construct the model based on the determined model type;

[0481] Building Main Structure and Ancillary Structure Model Construction: Based on the feature point coordinate information extracted from the array, the bottom elevation (h) of the main structure and ancillary structures is obtained from the bottom elevation calculation step. 底标高 The numerical values ​​for calculating the main building height and window height are obtained from the calculation steps. 主体高 The system uses numerical values ​​to draw building outlines and automatically constructs building models corresponding to feature point files of each building and its ancillary structures.

[0482] Sunlight window model construction: Based on the coordinate values ​​of feature points of each sunlight window on the same floor extracted from the array, the window sill elevation (h) obtained in the bottom elevation step is calculated. 底标高 The window height (h) is obtained by calculating the main body height and window height. 主体高 The numerical values ​​are used to extract the windowsill points of the sunshine windows, and one or more sunshine window models corresponding to each sunshine window feature point file are automatically constructed.

[0483] like Figure 6 As shown, the modeling method for a high-precision solar radiation analysis model based on point clouds includes the following steps in constructing the building main body and auxiliary structure models:

[0484] Drawing building outlines: Extract the (x, y) coordinates of each feature point belonging to the same block of the main building and its auxiliary structures from the array; automatically connect the feature points with outlines according to the coordinate order to generate the outer outline of the main building or the auxiliary structures; determine whether the outline is closed according to requirements; when it is determined that the outline needs to be closed, connect the last feature point to the first feature point end to end and set it as a closed outline; when it is determined that the outline does not need to be closed, the last feature point does not need to be connected to the first feature point.

[0485] like Figure 9 The image shows an example of drawing the outer contour line of a building by extracting the (x, y) coordinates of the building's main feature points from an array and connecting these feature points in sequence. Starting from ZT001, the contour line connects the feature points ZT002, ZT003, and ZT004 in a clockwise direction, ending at ZT004, and then the contour line is closed.

[0486] like Figure 10 The image shows an example of drawing the outer contour line of a building's auxiliary structure (balcony) by sequentially connecting the (x, y) coordinates of feature points extracted from an array. Starting from yt001, the contour line connects each feature point in a clockwise direction until yt004, at which point the contour line is closed, completing the drawing of the first balcony contour line. Starting from yt005, the contour line connects each feature point in a clockwise direction until yt008, at which point the contour line is closed, completing the drawing of the second balcony contour line. This process continues until the 3rd, 4th, ..., 10th balcony contour lines are drawn, until all 11 balconies are drawn.

[0487] like Figure 11 The image shows an example of drawing the outer contour line of a building's eaves by extracting the (x, y) coordinates of feature points from an array and connecting them sequentially. Starting from FY001, the contour line connects the feature points FY002, FY003, FY004, FY005, FY006, FY007, FY008, ..., FY035, FY036 in a clockwise direction, ending at FY036. The contour line is then closed.

[0488] Model construction: Set the bottom elevation value of the contour line to the bottom elevation (h) obtained from the bottom elevation calculation step. 底标高 The thickness value of the outline is set to the main body height (h) obtained from the steps of calculating the main body height and window height. 主体高 Set the bottom elevation value of the outline to the bottom elevation (h) obtained from the bottom elevation calculation step. 底标高After that, the block is modeled by stretching it up according to the outline. The height of the stretched block is the main body height (h) obtained from the steps of calculating the main body height and window height. 主体高 );

[0489] like Figure 12 The image shows an example of constructing a building body model based on the outline, bottom elevation, and main body height.

[0490] like Figure 13 The image shows an example of constructing a building attachment structure (balcony) model based on the outline, bottom elevation, and main body height.

[0491] like Figure 14 The image shows an example of constructing a building's auxiliary structure (eaves) model based on the outline, bottom elevation, and main body height.

[0492] When the outer contour shape of balconies on each floor is consistent and the spacing between balconies on different floors is within the tolerance range, the following method is adopted: first, a balcony model of a certain floor is established based on the contour line, bottom elevation, and main structure height; then, based on the feature point file information, the spacing value △H of the balconies in the Z direction is calculated, and the balconies of each floor are copied according to the spacing value array.

[0493] The calculation formula is:

[0494]

[0495] In the formula:

[0496] z m z m+1 This indicates the bottom elevation or top elevation of the balcony on the m-th or m+1-th floor;

[0497] n represents the number of balcony floors involved in the calculation and modeling;

[0498] △H represents the average spacing value.

[0499] like Figure 7 As shown, a high-precision solar radiation analysis modeling method based on point clouds is described, wherein the construction of the solar radiation window model includes the following steps:

[0500] Extracting left and right window sill points of sunlit windows: Extract the (x,y) coordinate values ​​of feature points of each sunlit window on the same floor from the array to distinguish between left and right window sill points;

[0501] Insert windows: Using the left and right endpoints of the window as the start and end points, insert windows at these two points to provide sunlight. Set the windowsill elevation to the windowsill elevation (h) obtained in the calculation of the bottom elevation step. 底标高 Set the window height to the height of the main body and the window height obtained in the window height calculation step (h). 主体高 );

[0502] like Figure 15As shown, the automatically constructed building main body, auxiliary structure model, and sunlight window model are placed in the corresponding three-dimensional space according to the coordinate values, and automatically combined to generate a complete high-precision sunlight analysis model.

[0503] like Figure 8 As shown, the high-precision solar radiation analysis modeling method based on point clouds includes the following steps for automatically constructing the hilltop model:

[0504] Retrieve slope crest feature point files: Based on the file name and storage path of the feature point files, retrieve all slope crest feature point files, and then iterate through and read the slope crest feature point files one by one;

[0505] Read feature point coordinates and store them in an array: Iterate through the feature point data in each feature point file line by line, obtain the coordinates (x, y, z) and remarks of each feature point in each feature point file, and store the coordinates and remarks of the feature points that constitute the top of the slope in the corresponding array in order; Perform coordinate unit conversion according to the steps of reading feature point coordinates and storing them in an array in the automatic construction of building main body, auxiliary structure and sunlight window model;

[0506] To create a 3D mesh surface, connect all feature points on the same slope point by point. The specific method is as follows: traverse the feature points in the array and extract them, connect them to construct a 3D surface, and calculate the slope and aspect of the 3D surface; calculate and obtain 3D surfaces with the same slope and aspect, extract the points that make up the 3D surface, and store their (x, y, z) values ​​in an array; iterate through the calculations, storing points on the same slope surface in the same array, resulting in a separate array for each slope surface; remove duplicate coordinates of the feature points (x, y, z) in the slope coordinate array; calculate the order of the coordinate points in each array, and connect the feature points on the slope surface with the same slope point one by one according to the order to generate a 3D mesh surface.

[0507] like Figure 16 The image shows point cloud data with a slope top, and a solar radiation model with a slope top generated based on the automatic slope top model building step.

[0508] In specific implementation cases, the ancillary structures include: balconies, eaves, columns, air conditioning partitions, unit partitions, moldings, sunshades, parapet walls of the roof, stairwells leading to the roof, water tank rooms, decorative components, and elevator machine rooms.

[0509] In specific implementation cases: the feature point information is stored in the following ways: feature point information of building main bodies and auxiliary structures with different bottom elevations and main body heights, feature point information of sun windows belonging to different floors, and feature point information of slopes not belonging to the same slope are saved to different feature point files. The feature point files are named according to the feature point type and are distinguished by adding suffixes, etc. The feature point information of building main bodies and auxiliary structures with different bottom elevations and main body heights, feature point information of sun windows belonging to different floors, and feature point information of slopes not belonging to the same slope are stored in the same file path according to category, and the folder is named according to the feature point type.

[0510] In specific implementation cases: the method of establishing a model based on the bottom elevation of the outline and the height of the main body includes: modeling by setting the bottom elevation of the outline and the height of the main body; or modeling by setting the bottom elevation of the outline and then raising the block, where the height of the raised block is the height of the main body.

[0511] Therefore, the high-precision solar radiation analysis modeling method based on point clouds in this embodiment of the invention can effectively solve the problem of rapidly constructing a high-precision solar radiation analysis model based on feature point information extracted from point cloud data. Compared with traditional modeling methods, the operation process is simplified, visualized, and intelligent, significantly reducing the workload and difficulty of field measurement and indoor modeling. By extracting feature points from point cloud data and storing them according to rules, the model can be directly generated, eliminating the need for traditional modeling methods to draw building outlines, calculate the height of various parts of the building, and manually input parameters for modeling. The requirements for the professional skills of the operators are also greatly reduced. Under the premise of ensuring the accuracy of the results, the efficiency of solar radiation analysis modeling is greatly improved, providing an efficient technical solution for establishing solar radiation analysis models of existing main and secondary buildings.

[0512] Although embodiments of the invention have been described by way of examples, those skilled in the art will recognize that the invention has many variations and modifications without departing from its spirit, and it is intended that the appended claims cover such variations and modifications without departing from the spirit of the invention.

Claims

1. A point cloud based high-fidelity daylight analysis model modeling method, characterized in that, The method comprises the following steps: Extract and store feature point information: obtain building point cloud data by using a three-dimensional laser scanner; perform point cloud data registration, coordinate system conversion, noise reduction and data thinning preprocessing operations on the building point cloud data; extract feature point information of the building according to rules, export the number and coordinate information of the feature points, and store the feature point files of the same type in the same file path according to the same rules; Automatic construction of a sunshine analysis model: traverse the feature point files, and automatically construct the building main body, the auxiliary structure model, the sunshine window model and the roof model corresponding to each feature point file based on the feature point type, the feature point number, the coordinate value and the note information; each single model is placed in the corresponding three-dimensional space according to the coordinate value, and the sunshine analysis model is automatically generated by combination; The extraction and storage of the feature point information comprises the following steps: The feature points of the building main body, the auxiliary structure, the sunshine window and the roof are extracted in the following manner: ； In the formula, U(s, a) represents a feature point extraction probability, c puct represents a constant of exploration level, P(s, a) represents a prior probability, N(s, b) represents a parent node access frequency, and N(s, a) represents a child node access frequency. In the extraction of feature point information constituting a building main body, an auxiliary structure, and a sunlight window, at least one lowermost feature point and one uppermost feature point are ensured. The main body height is calculated in the following manner: First, the coordinate values of the highest points are determined: z max = max(z r ), r = 1, 2,..., v; Secondly, the feature points participating in the calculation are screened according to the set threshold value Δz, if the s-th feature point z s The difference value of the coordinate value of the highest point is within the threshold value, and the calculation of the next step is participated in, wherein the judgment whether the feature point participates in the calculation of the next step is: z max -z s ≤Δz, s = 1, 2, …, u; Then, the top height is calculated: When the set threshold value Δz is 0, then h 顶部高 The calculation formula is: h 顶部高 = max(z r ), r = 1, 2,..., v; Finally, the subject height h is calculated: 主体高 = h 顶部高 - h 底标高 ; In the formula: h 主体高 is the main body height of the building main body, the main body height of the accessory structure, or the window height of the sunlight window; h 顶部高 is the top height of the building main body, the top height of the accessory structure, or the top height of the sunlight window; h 底标高 is the bottom elevation of the building main body, the bottom elevation of the accessory structure, or the bottom elevation of the sunlight window; z r is the coordinate value of the rth feature point, max is the maximum value, and v is the total number of feature points when calculating the main body height; z s represents the coordinate value of the sth top feature point within the threshold range participating in the calculation, and u represents the number of top feature points participating in the calculation.

2. The point cloud based high-fidelity daylight analysis model modeling method according to claim 1, wherein, The extraction and storage of the feature point information further comprises the following steps: The feature points of the building main body, the auxiliary structure, the sunshine window and the roof in the point cloud data are extracted to collect the coordinate information of the feature points; A sample input function is defined by the feature points of the building main body, the auxiliary structure, the sunshine window and the roof, the sample input function is used for training and evaluating the feature point extraction model, a hook callback function is defined and registered in the estimator framework to output detailed information in the training process, a for loop is established, and the training and evaluation of the feature point extraction model are performed in the loop, and relevant information is output; after the training is completed, the feature point extraction model is exported; According to the feature point extraction model, the feature points of the building main body, the auxiliary structure, the sunshine window and the roof are automatically extracted; the automatically extracted feature points are checked, and the feature points with problems are corrected by human-computer interaction; the building main body feature points and the auxiliary structure feature points belonging to the same block, the sunshine window feature points of the same floor, and the feature points constituting the same roof are randomly determined as starting points, and then the feature points are numbered in sequence, and whether the note information needs to be filled in is determined; the building main body feature points belonging to the same block, the building auxiliary structure feature points belonging to the same block, the sunshine window feature points of the same floor with the same window sill bottom height and window height, and the feature points constituting the same roof are saved as separate feature point files.

3. The point cloud based high-fidelity daylight analysis model modeling method of claim 1, wherein: The automatic construction of the sunshine analysis model comprises the following steps: Automatic construction of the building main body, the auxiliary structure and the sunshine window model: retrieve all feature point files of the building main body, the auxiliary structure and the sunshine window stored in the specified path and named according to the preset naming manner, and read each feature point file one by one; Traverse the feature point data in each feature point file line by line, and store the feature point data in each feature point file in different arrays; Traverse the feature point data in each feature point file line by line, and store the feature point data in each feature point file in different arrays; The bottom elevation and the main body height of each building main body and accessory structure, and the bottom elevation and the window height of each floor or each sunlight window are calculated based on the z coordinate values of the feature points in different arrays; The contour line is connected based on the coordinate values of the feature points extracted from the array and belonging to the same block to generate the outer contour line of the building main body or the outer contour line of the accessory structure, and the model of the building main body and the accessory structure is established based on the bottom elevation and the main body height of the contour line; The coordinate values of the left and right window sill feature points of each sunlight window of the same floor are extracted from the array, and the sunlight window model is established by using the two-point window insertion method based on the coordinate values of the window sill feature points, the bottom elevation and the window height. The automatic construction of the roof model includes the following steps:

4. The point cloud based high-fidelity daylight analysis model modeling method of claim 3, wherein: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: First, the coordinate values of the minimum points are determined: z min = min(z i ), i = 1, 2, …, w; Secondly, the feature points participating in the calculation are screened according to the set threshold value Δh, if the jth feature point z j The difference value of the coordinate value of the lowest point is within the threshold value, then the calculation of the next step is participated in, wherein the judgment of whether the feature point participates in the calculation of the next step is: z j -z min ≤Δh, j = 1, 2, …, m; Finally, the bottom elevation is calculated: When the threshold value Ah is set to 0, then h 底标高 The calculation formula is: h 底标高 = min(z i ), i = 1, 2, …, w; In the formula: h 底标高 is the bottom elevation of the main building, the bottom elevation of the auxiliary structure, or the bottom elevation of the solar window; z i is the coordinate value of the i-th feature point, min is the minimum value, and w is the total number of feature points when calculating the bottom elevation; z j represents the coordinate value of the j-th bottom feature point within the threshold range participating in the calculation; m represents the number of bottom feature points participating in the calculation; The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps:

5. The point cloud based high-fidelity solar analysis model modeling method of claim 3, wherein, The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the building main body, the accessory structure and the sunlight window model includes the following steps: The automatic construction of the Extracting the coordinate values of the feature points belonging to the same building main body and accessory structure from the array respectively; automatically connecting the feature points using the contour line according to the coordinate point order to generate the outer contour line of the building main body or the outer contour line of the accessory structure; determining whether the contour line is closed according to the requirement; connecting the last feature point with the first feature point when it is determined that the contour line needs to be closed, and setting it as a closed contour line; when it is determined that the contour line does not need to be closed, the last feature point does not need to be connected with the first feature point; Based on the coordinate values of the feature points belonging to the same building main body and accessory structure in the array, automatically calculating the bottom elevation and the main body height; setting the bottom elevation value of the contour line as the bottom elevation; then setting the thickness value of the contour line as the main body height to complete the model construction, or modeling according to the contour line to complete the model construction, and the height value of the body block is the main body height; When the balcony outer contour shape of each floor is consistent, and the interval value between the balconies of different floors is within the preset range, a balcony model of a certain floor is established according to the contour line, the bottom elevation and the main body height, then the interval value ΔC of the balcony in the Z direction is calculated according to the feature point file information, and the balconies of each floor are copied according to the interval value array, The calculation formula is: △C= ； In the formula: z q , z q+1 respectively represent the bottom elevation and top elevation of the qth and q+1th balcony; t represents the number of balconies participating in the calculation; and ΔC represents the average interval value.

6. The point cloud based high-fidelity solar analysis model modeling method of claim 4, wherein, The sunlight window model construction includes the following steps: Extracting the coordinate values of the feature points of each sunlight window of the same floor from the array, and distinguishing the left and right window sill points; Based on the coordinate values of the feature points of the sunlight window, calculating the window height of the sunlight window and the bottom elevation of the sunlight window; setting the window sill elevation value as the bottom elevation of the sunlight window, and setting the window height value as the window height of the sunlight window; then taking the left and right end points of the window as the starting point and the ending point of the window, and using the two-point window insertion method to insert the windows one by one to complete the construction of the model of each sunlight window of the same floor.

7. The point cloud based high-fidelity solar analysis model modeling method of claim 3, wherein: The automatic construction of the slope roof model includes the following steps: Retrieving the slope feature point file: retrieving all the slope feature point files according to the file naming and storage path of the feature point file, and then traversing and reading the slope feature point file one by one; Reading the feature point coordinates and storing them in the array: traversing the feature point data in each feature point file line by line, obtaining the coordinates and note information of each feature point in each feature point file, and storing the feature point coordinates and note information constituting the slope roof in the corresponding array in order; Drawing a three-dimensional mesh surface: connecting each feature point on the same slope surface point by point to generate a three-dimensional mesh surface.

8. The point cloud based high-fidelity solar analysis model modeling method according to any one of claims 1-7, characterized in that: The accessory structure includes: balcony, eaves, column, air conditioner partition, household partition, line foot, sunshade, daughter wall of roof, out-of-roof stairwell, water tank room, decorative component and elevator machine room; the feature point information of the building main body and accessory structure with different bottom elevations and main body heights, the feature point information of the sunlight window belonging to different floors, and the slope feature point information not belonging to the same slope roof are stored in the same file path according to the category, and the folder is named according to the feature point type; the model construction method based on the bottom elevation and the main body height of the contour line includes: modeling by setting the bottom elevation and the main body height of the contour line, or modeling by setting the bottom elevation of the contour line and then lifting the body block.

Images