Building core tube generation method and system based on data feature adaptation

Abstract

The invention discloses a building core tube generation method and system based on data feature adaptation, and the method comprises the steps: obtaining case sample data, constructing a structure database according to the case sample data, obtaining building data, extracting building data features, and calculating the similarity between the building data features and the case sample data in the structure database, and selecting a plurality of case samples as alternative samples according to the similarity, merging the building data with each alternative sample to obtain alternative arrangement schemes, rechecking each alternative arrangement scheme, and selecting a recommended arrangement scheme from the alternative arrangement schemes. According to the method, the building samples are obtained and the database is constructed, so that similarity analysis and matching are carried out by utilizing the database, the candidate schemes are selected and re-checked and screened, and the problems that an existing high-rise building core tube is high in design complexity and cannot be subjected to multi-target collaborative optimization are solved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of architectural design, and in particular to a building core tube generation method and system based on data feature adaptation. BACKGROUND

[0002] The design of a high-rise building core tube needs to consider multiple objectives such as building function layout, structural mechanical performance, and comprehensive mechanical and electrical pipeline, etc. However, the existing design process highly depends on the individual experience of designers, who need to manually compare historical case data, which takes several weeks and is prone to missing key constraint conditions. In the traditional forward design process, the building scheme at the initial stage often only focuses on the plane layout and spatial form, without fully embedding the equipment pipeline path and the boundary conditions of the structural reinforcement zone, which leads to forced and revolutionary adjustments when the mechanical and electrical and structural professionals intervene later. For example, the position offset of the equipment shaft may cause the weakening of the core tube shear wall opening, which requires rechecking the overall stiffness of the structure and supplementing the structural edge components. Such a "design-correction" cycle not only reduces efficiency, but also may lead to the failure of standard review, increasing the construction risk.

[0003] However, the existing plane generation algorithm faces significant bottlenecks in core tube design. Although genetic algorithms have global search capabilities, their binary coding method is difficult to directly map the nonlinear constraints of the core tube, which is prone to local optimal solutions and has poor dynamic adaptation to standards, difficulty in multi-objective optimization, and low human-computer collaboration efficiency. SUMMARY

[0004] To solve the above problems, the application provides a building core tube generation method and system based on data feature adaptation, which acquires building samples and constructs a database to perform similarity analysis and matching using the database to select candidate schemes and review and screen, solving the problem of high complexity of existing high-rise building core tube design and inability to multi-objective collaborative optimization.

[0005] To achieve the above purpose, the application provides the following technical solutions: A building core tube generation method based on data feature adaptation, comprising the following steps: S1. Acquiring case sample data and constructing a structure database according to the case sample data; S2. Acquiring building data, extracting building data features, calculating the similarity between the building data features and the case sample data in the structure database, and selecting a plurality of case samples as candidate samples according to the similarity; S3. Merging the building data with each candidate sample to obtain a candidate arrangement scheme, reviewing each candidate arrangement scheme, and selecting a recommended arrangement scheme from the candidate arrangement scheme.

[0006] Further, in step S1, the case sample data is acquired, and a structure database is constructed according to the case sample data, and a specific implementation manner is as follows: Obtain drawings of a plurality of different case samples, and classify the case samples according to function types; For each type of case sample, record the building parameters of each case sample, calculate the core tube utilization rate of the case sample according to the building parameters of each case sample, and screen the case sample according to the core tube utilization rate, store the screened case sample data into the structure database according to the function types, and map each case sample data to the corresponding case sample drawing.

[0007] Further, the building parameters of each case sample are recorded respectively, the core tube utilization rate of the case sample is calculated according to the building parameters of each case sample, and the case sample is screened according to the core tube utilization rate, and a specific implementation manner is as follows: The building parameters include the standard layer area, the number of tower stories, the standard layer width, the standard layer depth, the core tube length and the core tube width of the case sample, and the calculation formula of the core tube utilization rate is as follows:

[0008] Wherein, I is the core tube utilization rate, A is the standard layer area, E is the core tube length, and F is the core tube width; For each type of case sample, grouping is performed according to the number of tower stories, the core tube utilization rates of the case samples in each group are compared, and the case samples with a core tube utilization rate lower than the average value in the group are excluded.

[0009] Further, in step S2, the building data is acquired, and the building data features are extracted, and a specific implementation manner is as follows: The building data includes the function type, the standard layer geometric contour and the number of tower stories, the geometric contour area and the convex hull of the geometric contour are calculated according to the standard layer geometric contour, the standard layer width and the standard layer depth are calculated based on the convex hull of the geometric contour, and the geometric contour area, the standard layer width, the standard layer depth and the number of tower stories are constructed as the building data feature vector.

[0010] Further, in step S2, the similarity between the building data features and the case sample data in the structure database is calculated, and a specific implementation manner is as follows: The case sample data with the same function type as the building data is taken out from the structure database as a comparison sample, for each comparison sample, the deviation value of each dimension data of the building data feature vector and the corresponding dimension data of the comparison sample is calculated, and a specific calculation formula is as follows:

[0011] Wherein, The range is 1-4. This represents the deviation value in the i-th dimension. This represents the i-th dimension of the feature vector of building data. This indicates the i-th dimension of the comparison sample; Calculate the weighted Euclidean distance between the feature vector of the building data and the comparison sample. The specific formula for calculating the weighted Euclidean distance is as follows:

[0012] Where D represents the weighted Euclidean distance between the feature vector of the building data and the comparison sample. The deviation value representing the area of ​​the geometric contour. The weights representing the area of ​​the geometric contour. This represents the deviation value of the standard plane width. This indicates the weight of the standard layer. This represents the deviation value of the standard layer depth. The weights representing the standard layer depth This indicates the deviation value of the number of floors in the tower. This indicates the weight of the tower floor number.

[0013] Furthermore, in step S2, the specific implementation of selecting several case samples as candidate samples based on similarity is as follows: sorting them according to the weighted Euclidean distance between the building data feature vector and the comparison samples, and selecting the top n comparison samples with the smallest weighted Euclidean distance as candidate samples.

[0014] Furthermore, in step S3, the process of merging the building data with each candidate sample to obtain candidate layout schemes is specifically implemented as follows: Each candidate sample is converted into a BIM model to obtain several core tube models. The geometric contours of the standard floors are then merged with each core tube model group to obtain the candidate layout schemes.

[0015] Furthermore, in step S3, the specific implementation method for reviewing each alternative layout scheme according to the specification clauses and calculating the deviation value of each alternative layout scheme is as follows: Fire evacuation verification of alternative layout schemes includes: calculating the total tower building area based on the standard floor area and number of floors of the alternative layout schemes; calculating the total number of evacuees based on the preset ratio of people per unit area; calculating the total evacuation width based on the total number of evacuees; comparing the total evacuation width with the staircase width of the core tube of the alternative layout scheme; if the staircase width of the core tube of the alternative layout scheme is less than the total evacuation width, the alternative layout scheme is eliminated. The fire distance of the alternative arrangement scheme is reviewed, including: calculating the farthest path distance between the core tube stairwell door and the standard floor contour of the alternative arrangement scheme, and if the path distance is greater than the preset distance, the alternative arrangement scheme is eliminated; The elevator quantity of the alternative arrangement scheme is reviewed, including: calculating the elevator configuration quantity according to the total tower building area, comparing the elevator configuration quantity with the elevator quantity of the alternative arrangement scheme, and recording the deviation value, The deviation value = elevator configuration quantity - elevator quantity of the alternative arrangement scheme.

[0016] Further, in step S3, the recommended arrangement scheme is selected from the alternative arrangement schemes, and the specific implementation manner is: According to the deviation value of each alternative arrangement scheme, the alternative arrangement schemes are sorted from small to large, and the alternative arrangement scheme with the smallest deviation value is selected as the recommended arrangement scheme, and if there are multiple alternative arrangement schemes with the smallest deviation value, the alternative arrangement scheme with the smallest weighted Euclidean distance is selected as the recommended arrangement scheme.

[0017] Through the above technical scheme, the present application has the following beneficial effects: Deeply combine the characteristics of the building industry, focus on breaking through the design difficulties of the core tube of the high-rise building, form a full-process automatic solution scheme from scheme generation, key data verification to main node construction drawing output, which can be applied to the automatic design requirements of the core tube of high-rise buildings of various heights, and quickly provide the most matched core tube arrangement scheme for designers from multiple indicators. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 It is a whole process schematic diagram of a building core tube generation method based on data feature adaptation.

[0019] Figure 2 It is a structure schematic diagram of a building core tube generation system based on data feature adaptation in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0021] In order to make the above-mentioned purposes, characteristics and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0022] Example 1 Referring to Figure 1 A building core tube generation method based on data feature adaptation, comprising the following steps: S1. Obtain case sample data, and construct a structure database according to the case sample data; S2. Obtain building data, extract building data features, calculate the similarity between the building data features and the case sample data in the structure database, and select a plurality of case samples as candidate samples according to the similarity; S3. Merge the building data with each candidate sample respectively to obtain candidate arrangement schemes, review each candidate arrangement scheme, and select a recommended arrangement scheme from the candidate arrangement schemes.

[0023] In an optional embodiment, in step S1, the case sample data is obtained, and the structure database is constructed according to the case sample data, and the specific implementation manner is: Obtain drawings of a plurality of different case samples, and classify the case samples according to function types; For each type of case sample, record the building parameters of each case sample, calculate the core tube utilization rate of the case sample according to the building parameters of each case sample, and screen the case samples according to the core tube utilization rate, store the screened case sample data in the structure database according to the function types, and map each case sample data to the corresponding case sample drawing.

[0024] The core tube of a high-rise building is a vertical structural system located inside the building, integrating functions such as elevators, staircases, pipe wells, etc., which not only bears the vertical and horizontal loads of the building and provides structural stability, but also concentrates the arrangement of mechanical and electrical equipment and evacuation passages, and is a "central system" in high-rise buildings that combines structure and function, which is crucial to the safety, functionality and spatial organization of the building.

[0025] Specifically, the function types include office, medical, hotel, etc., and the different case samples include built cases of different standard layer morphologies and numbers of layers, and specifically, the following data is recorded for each case sample: standard layer area, total number of tower floors, standard layer width, standard layer depth, core tube length, core tube width, number of elevator arrangements, width of the lobby, and equipment room area. Each case sample in the structure database is assigned a sample serial number, which is mapped to the corresponding electronic drawing for quick retrieval of the arrangement scheme, thus completing the construction of the case structured database and supporting the addition and import of new samples.

[0026] In an optional embodiment, the building parameters of each case sample are recorded, the core tube utilization rate of the case sample is calculated according to the building parameters of each case sample, and the case samples are screened according to the core tube utilization rate, and the specific implementation manner is: The building parameters include the standard floor area, the number of tower stories, the standard floor width, the standard floor depth, the core tube length, and the core tube width of the case sample, and the calculation formula of the core tube utilization rate is:

[0027] Wherein, I is the core tube utilization rate, A is the standard floor area, E is the core tube length, and F is the core tube width. For each type of case sample, grouping is performed according to the number of tower stories, and the core tube utilization rates of case samples in each group are compared, and case samples with a core tube utilization rate lower than the average value in the group are screened out.

[0028] Specifically, and according to the number of tower stories from low to high, every 5 stories is a comparison group, the average core tube utilization efficiency of case samples in the group is calculated, and the specific way of calculating the average core tube utilization efficiency of case samples in the group further includes calculating the median of case samples in the group.

[0029] In an optional embodiment, in step S2, building data is obtained, and building data features are extracted, and the specific implementation manner is: The building data includes the function type, the standard floor geometric contour, and the number of tower stories, the geometric contour area and the convex hull of the geometric contour are calculated according to the standard floor geometric contour, the standard floor width and the standard floor depth are calculated based on the convex hull of the geometric contour, and the geometric contour area, the standard floor width, the standard floor depth, and the number of tower stories are constructed as a building data feature vector.

[0030] The building data is obtained by obtaining the scheme basis information input by the user, specifically, the convex hull of the geometric contour is calculated by the AABB algorithm: the contour point coordinates are extracted, the convex hull point set is calculated by using the convex hull algorithm, the coordinate extreme values are extracted from the convex hull, and the axis-aligned rectangle defined by the extreme values, i.e. the convex hull, is constructed. The convex hull is a smallest rectangle or cuboid that is aligned on the Cartesian coordinate axis and completely contains the target object. The principal axis direction is calculated according to the convex hull, and the standard floor width, the standard floor depth, and the perimeter of the geometric contour are finally determined, and the area or volume is calculated for subsequent geometric analysis.

[0031] In an optional embodiment, in step S2, the similarity between the calculated building data features and the case sample data in the structure database is calculated, and the specific implementation manner is: The case sample data with the same function type as the building data is taken out from the structure database as a comparison sample, for each comparison sample, the deviation value of each dimension data of the building data feature vector and the corresponding dimension data of the comparison sample is calculated, and the specific calculation formula is:

[0032] wherein, the range of w is 1-4, represents the deviation value of the i-th dimension, represents the i-th dimension data of the building data feature vector, represents the i-th dimension data of the contrast sample; The weighted Euclidean distance of the building data feature vector and the contrast sample is calculated, and the specific calculation formula of the weighted Euclidean distance is:

[0033] wherein, D represents the weighted Euclidean distance of the building data feature vector and the contrast sample, represents the deviation value of the geometric profile area, represents the weight of the geometric profile area, represents the deviation value of the standard layer width, represents the weight of the standard layer width, represents the deviation value of the standard layer depth, represents the weight of the standard layer depth, represents the deviation value of the number of tower stories, represents the weight of the number of tower stories.

[0034] Specifically, the geometric profile area, the standard layer width, the standard layer depth and the number of tower stories of each contrast sample are extracted to form a matrix M of n*4, wherein n is the number of contrast samples, and the absolute proportion of the deviation value of each dimension of the building data feature vector and each sample of the matrix M is calculated.

[0035] Since the core tube design is sensitive to different dimensions of data, the weights of each dimension are also different. Generally, the area and the number of stories have a greater impact on the scheme, so the corresponding dimension weight value is higher. Specifically, the area weight w1=0.375, the surface width weight w2=0.125, the depth weight w3=0.125, and the number of stories weight w4=0.375. However, this weight parameter is the initial calculation weight of the algorithm, and in actual operation, the loss is calculated according to the sample input and output results, and the weight is adjusted. After the weighted Euclidean distance is calculated, the sample cases are finally sorted according to the weighted Euclidean distance, and the six sample cases with the smallest distance are selected, and their serial numbers are recorded for subsequent retrieval of the layout scheme.

[0036] In an optional embodiment, in step S2, the several case samples are selected as the candidate samples according to the similarity, and the specific implementation manner is that the weighted Euclidean distances between the building data feature vector and the contrast samples are sorted, and the first n contrast samples with the smallest weighted Euclidean distance are selected as the candidate samples.

[0037] In an optional embodiment, in step S3, the merging of the building data with each of the candidate samples respectively to obtain the candidate arrangement schemes is specifically implemented as follows: Each of the candidate samples is converted into a BIM model to obtain a plurality of core tube models, and the standard layer geometric profile is merged with each of the core tube models to obtain the candidate arrangement schemes.

[0038] Based on the IFC standard, the candidate samples are converted into a parameterized BIM model, the information of the corresponding functional space size of the sample case is read according to the electronic drawing corresponding to the candidate sample and is reconstructed in the BIM system, the script code corresponding to the wall generation command is generated, the core tube wall, equipment room, elevator shaft and other components are automatically filled to obtain the core tube model, and the core tube model is spliced according to the standard layer geometric profile of the building data. Specifically, the standard layer geometric profile input by the user is a regular rectangular profile, and the core tube model is aligned with the geometric center of the rectangular profile and then spliced.

[0039] In an optional embodiment, in step S3, the checking of each of the candidate arrangement schemes according to the specification clauses to calculate the deviation value of each of the candidate arrangement schemes is specifically implemented as follows: The candidate arrangement schemes are checked for fire evacuation, including: calculating the total tower building area according to the standard layer area and the number of layers of the candidate arrangement schemes, calculating the total evacuation number according to the preset unit area person ratio, calculating the total evacuation width according to the total evacuation number, and comparing the total evacuation width with the stair width of the core tube of the candidate arrangement scheme. If the stair width of the core tube of the candidate arrangement scheme is less than the total evacuation width, the candidate arrangement scheme is removed. The candidate arrangement schemes are checked for fire distance, including: calculating the farthest path distance between the core tube stairwell door and the standard layer profile of the candidate arrangement scheme, and removing the candidate arrangement scheme if the path distance is greater than the preset distance. The candidate arrangement schemes are checked for the number of elevators, including: calculating the elevator configuration number according to the total tower building area, comparing the elevator configuration number with the number of elevators of the candidate arrangement scheme, and recording the deviation value, The deviation value = elevator configuration number - number of elevators of the candidate arrangement scheme.

[0040] In actual application, the review indicators are distributed according to the requirements of the fire safety regulations, wherein, 5000 square meters per part of elevators are allocated for tower heights below 100 meters, and 10000 square meters per part of elevators are allocated for tower heights of 100 meters or above.

[0041] In an optional embodiment, in step S3, the recommended arrangement scheme is selected from the candidate arrangement schemes, and the specific implementation is as follows: According to the deviation values of each alternative arrangement scheme, the alternative arrangement scheme with the smallest deviation value is selected as the recommended arrangement scheme from small to large, and if there are multiple alternative arrangement schemes with the smallest deviation value, the alternative arrangement scheme with the smallest weighted Euclidean distance is selected as the recommended arrangement scheme.

[0042] In an optional embodiment, the specification verification is performed synchronously in the scheme generation stage, the non-compliant items are automatically labeled and the optimization strategy is recommended, in addition, a plurality of candidate schemes passing the verification can be provided, and the weighted Euclidean distance specific value is given as a matching score for the designer to refer to, to ensure that it can be used for subsequent deepening design.

[0043] In an optional embodiment, a plug-in interface with Revit, AutoCAD and other software is developed to support users to call functions in the original design environment.

[0044] Example 2 Referring to Figure 2 A building core tube generation system based on data feature adaptation, comprising: A data construction module configured to obtain case sample data and construct a structure database according to the case sample data; A sample alternative module configured to obtain building data, extract building data features, calculate the similarity between the building data features and the case sample data in the structure database, and select a plurality of case samples as alternative samples according to the similarity; A scheme recommendation module configured to merge the building data with each alternative sample to obtain alternative arrangement schemes, review each alternative arrangement scheme, and select a recommended arrangement scheme from the alternative arrangement schemes. The embodiments disclosed in the specification are only an example of one-sided features of the present application, and the protection scope of the present application is not limited to this embodiment. Any other functionally equivalent embodiments fall within the protection scope of the present application. For those skilled in the art, other various corresponding changes and modifications can be made according to the above described technical solutions and concepts, and all these changes and modifications should belong to the protection scope of the claims of the present application.

Claims

1. A building core tube generation method based on data feature adaptation, characterized in that, The method comprises the following steps: S1. Obtain case sample data, and construct a structure database according to the case sample data; S2. Obtain building data, extract building data features, calculate the similarity between the building data features and the case sample data in the structure database, and select a plurality of case samples as candidate samples according to the similarity; S3. Merge the building data with each candidate sample to obtain a candidate arrangement scheme, review each candidate arrangement scheme, and select a recommended arrangement scheme from the candidate arrangement schemes.

2. The method of claim 1, wherein, In step S1, the case sample data is obtained, and the structure database is constructed according to the case sample data, and the specific implementation manner is: Obtain drawings of a plurality of different case samples, and classify the case samples according to the functional types; For each type of case sample, record the building parameters of each case sample, calculate the core tube utilization rate of the case sample according to the building parameters of each case sample, and screen the case samples according to the core tube utilization rate, store the screened case sample data in the structure database according to the functional types, and map each case sample data to the corresponding case sample drawing.

3. The method of claim 2, wherein, Record the building parameters of each case sample, calculate the core tube utilization rate of the case sample according to the building parameters of each case sample, and screen the case samples according to the core tube utilization rate, and the specific implementation manner is: The building parameters include the standard layer area, the number of tower stories, the standard layer width, the standard layer depth, the core tube length and the core tube width of the case sample, and the calculation formula of the core tube utilization rate is: Wherein, I is the core tube utilization rate, A is the standard layer area, E is the core tube length, and F is the core tube width; For each type of case sample, group according to the number of tower stories, compare the core tube utilization rates of the case samples in each group, and exclude the case samples with core tube utilization rates lower than the average value in the group.

4. The method of claim 3, wherein, In step S2, the building data is obtained, and the building data features are extracted, and the specific implementation manner is: The building data includes the functional type, the standard layer geometric contour and the number of tower stories, the geometric contour area and the convex hull of the geometric contour are calculated according to the standard layer geometric contour, the standard layer width and the standard layer depth are calculated based on the convex hull of the geometric contour, and the geometric contour area, the standard layer width, the standard layer depth and the number of tower stories are constructed into a building data feature vector.

5. The method of claim 4, wherein, In step S2, the similarity between the building data features and the case sample data in the structure database is calculated, and the specific implementation manner is: Take the case sample data with the same functional type as the building data from the structure database as the comparison sample, for each comparison sample, calculate the deviation value of each dimension data of the building data feature vector and the corresponding dimension data of the comparison sample, and the specific calculation formula is: wherein, ranging from 1 to 4, represents the deviation value of the i-th dimension, represents the i-th dimensional data of the building data feature vector, represents the i-th dimensional data of the contrast sample; Calculate the weighted Euclidean distance between the building data feature vector and the comparison sample, and the specific calculation formula of the weighted Euclidean distance is: D represents a weighted Euclidean distance between the building data feature vector and the contrast sample, represents a deviation value of the geometric profile area, represents a weight of the geometric profile area, represents a deviation value of the standard layer width, represents a weight of the standard layer width, represents a deviation value of the standard layer depth, represents a weight of the standard layer depth, represents a deviation value of the number of tower stories, represents a weight of the number of tower stories.

6. The method of claim 5, wherein, In step S2, the several case samples are selected as the candidate samples according to the similarity, and a specific implementation manner is as follows: the weighted Euclidean distances between the building data feature vectors and the comparison samples are sorted, and the first n comparison samples with the smallest weighted Euclidean distances are selected as the candidate samples.

7. The method of claim 6, wherein, In step S3, the building data is combined with each candidate sample respectively to obtain a candidate arrangement scheme, and a specific implementation manner is as follows: Each candidate sample is converted into a BIM model to obtain several core tube models, and the standard layer geometric profile is combined with each core tube model group to obtain a candidate arrangement scheme.

8. The method of claim 7, wherein, In step S3, each candidate arrangement scheme is reviewed according to the specification clauses, and a deviation value of each candidate arrangement scheme is calculated, and a specific implementation manner is as follows: The candidate arrangement scheme is reviewed for fire evacuation, including: the total tower building area is calculated according to the standard layer area and the number of layers of the candidate arrangement scheme, the total evacuation number is calculated according to the preset unit area person ratio, the total evacuation width is calculated according to the total evacuation number, and the total evacuation width is compared with the stair width of the core tube of the candidate arrangement scheme. If the stair width of the core tube of the candidate arrangement scheme is smaller than the total evacuation width, the candidate arrangement scheme is removed. The candidate arrangement scheme is reviewed for fire distance, including: the farthest path distance between the core tube stair door of the candidate arrangement scheme and the standard layer profile is calculated, and if the path distance is greater than the preset distance, the candidate arrangement scheme is removed. The number of elevators of the candidate arrangement scheme is reviewed, including: the elevator configuration number is calculated according to the total tower building area, the elevator configuration number is compared with the number of elevators of the candidate arrangement scheme, and the deviation value is recorded, The deviation value = the elevator configuration number - the number of elevators of the candidate arrangement scheme.

9. The method of claim 8, wherein, In step S3, the recommended arrangement scheme is selected from the candidate arrangement schemes, and a specific implementation manner is as follows: According to the deviation value of each candidate arrangement scheme, the candidate arrangement schemes are sorted from small to large, the candidate arrangement scheme with the smallest deviation value is selected as the recommended arrangement scheme, and if there are multiple candidate arrangement schemes with the smallest deviation value, the candidate arrangement scheme with the smallest weighted Euclidean distance is selected as the recommended arrangement scheme from the candidate arrangement schemes with the smallest deviation value.

10. A building corewall generation system based on data feature adaptation, characterized by, The data construction module is configured to obtain case sample data and construct a structure database according to the case sample data. The sample candidate module is configured to obtain building data, extract building data features, calculate the similarity between the building data features and the case sample data in the structure database, and select several case samples as candidate samples according to the similarity. The scheme recommendation module is configured to combine the building data with each candidate sample respectively to obtain a candidate arrangement scheme, review each candidate arrangement scheme, and select a recommended arrangement scheme from the candidate arrangement schemes. ​

Images