Building group model simulation construction method based on GIS+BIM

Abstract

The application relates to the technical field of data processing, in particular to a building group model simulation construction method based on GIS+BIM, which comprises the following steps: according to the moving matching condition between the coincident part and the difference part of the point cloud data between every two monomer buildings, and in combination with the point cloud distribution of the difference part, the comprehensive effectiveness between every two monomer buildings is obtained; the monomer buildings are divided by using the comprehensive effectiveness to obtain each basic building group; according to the distance distribution between the point cloud data of the difference part between every two monomer buildings in the basic building group, the structural degree of the point cloud data of the difference part is analyzed to obtain the basic necessity; the basic building group is divided into each building group; according to each building group in each basic building group, a modeling scheme of a shared basic BIM model is formulated, and GIS is combined to simulate the building group model. The application guarantees the accuracy of model construction and the efficiency of model construction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a building group model simulation construction method based on GIS+BIM. BACKGROUND

[0002] With the increasing demand for fine construction of building group models, building group models not only need to meet the requirements of three-dimensional visualization display, but also need to have the simulation capability of building space relationship, attribute information and dynamic process. In the prior art, geographic information system (Geographic Information System or Geo-Information system, GIS) mainly focuses on the management and analysis of macro geographic spatial data, while building information modeling (Building Information Modeling, BIM) focuses on the fine expression and component-level information management of single buildings. Using the combination of GIS and BIM to construct building group models can ensure the accuracy of the overall space and take into account the detailed expression of single buildings in building group simulation, thereby providing more comprehensive, detailed and dynamic digital support for urban planning, construction management and smart city applications.

[0003] In the prior art, in order to meet the fine construction requirements of BIM models, modeling is mainly performed through architectural CAD drawings to obtain the BIM model of the building group. However, when the scale of the building group is large, each building usually needs to be modeled one by one, resulting in a large amount of repetitive work, a long modeling period and low modeling efficiency. SUMMARY

[0004] In order to solve the technical problem of low efficiency of the prior art method which needs to model each building one by one, the purpose of the present application is to provide a building group model simulation construction method based on GIS+BIM, and the technical solution adopted is as follows:

[0005] Obtain point cloud data of each single building;

[0006] According to the movement matching condition between the overlapping part and the difference part of the point cloud data between each two single buildings, and combining the point cloud distribution of the difference part, obtain the comprehensive effectiveness between each two single buildings; divide the single buildings by using the comprehensive effectiveness to obtain each basic building group;

[0007] According to the distance distribution between the point cloud data of the difference part between each two single buildings in the basic building group, analyze the structural degree of the point cloud data of the difference part, and obtain the basic necessity between each two single buildings in the basic building group;

[0008] Dividing the basic building groups into each building group according to the basic necessity; according to each building group in each basic building group, making a modeling scheme of a shared basic BIM model, and combining GIS to simulate a building group model.

[0009] Preferably, the comprehensive effectiveness between the two monomer buildings is obtained according to the movement matching condition between the coincident part and the difference part of the point cloud data between the two monomer buildings, and the point cloud distribution of the difference part, and specifically includes:

[0010] Taking any two different monomer buildings as a first building and a second building respectively; matching the two monomer buildings to obtain a coincident point cloud set and a non-coincident point cloud set between the first building and the second building; the non-coincident point cloud set includes several difference point cloud subsets of the first building or the second building;

[0011] Obtaining a basic inhibition index of each difference point cloud subset according to the data matching condition between each difference point cloud subset and the coincident point cloud set;

[0012] Obtaining an invalidity characteristic factor of each monomer building according to the number of point cloud data contained in each difference point cloud subset corresponding to each monomer building and the basic inhibition index;

[0013] Performing negative correlation normalization processing on the maximum value of the invalidity characteristic factors of the first building and the second building to obtain the comprehensive effectiveness between the first building and the second building.

[0014] Preferably, the basic inhibition index of each difference point cloud subset is obtained according to the data matching condition between each difference point cloud subset and the coincident point cloud set, and specifically includes:

[0015] Taking the direction from the center point cloud of each difference point cloud subset to the center point cloud of the coincident point cloud set as the movement matching direction of each difference point cloud subset;

[0016] Moving each difference point cloud subset according to a preset movement step and the movement matching direction, and after each movement, matching the difference point cloud subset and the coincident point cloud set by using a matching algorithm, and taking the proportion of the point cloud data matched successfully after each movement as the basic matching degree of each difference point cloud subset under each movement;

[0017] Determining the basic inhibition index of each difference point cloud subset based on the variance and mean value of the basic matching degree corresponding to each difference point cloud subset under all movements.

[0018] Preferably, the number of point cloud data contained in each difference point cloud subset corresponding to each single building and the basic inhibitory index are used to obtain a failure characteristic factor of each single building, and specifically include:

[0019] For any single building in the first building or the second building, a failure characteristic factor of the single building is determined based on a product of a normalized result of the number of all point cloud data in each difference point cloud subset and the basic inhibitory index.

[0020] Preferably, the structural similarity of the point cloud data of the difference part between each two single buildings in the basic building group is analyzed based on the distance distribution between the point cloud data of the difference part between each two single buildings in the basic building group, to obtain a basic necessity between each two single buildings in the basic building group, and specifically includes:

[0021] The structural similarity between each two difference point cloud subsets is obtained based on the distance distribution between the point cloud data matched by the different difference point cloud subsets between each two single buildings in the basic building group.

[0022] All difference point cloud subsets between each two single buildings in the basic building group are classified based on the structural similarity to obtain a plurality of common structure clusters between each two single buildings.

[0023] The basic necessity between each two single buildings in the basic building group is obtained based on the number of difference point cloud subsets in each common structure cluster between each two single buildings in the basic building group.

[0024] Preferably, the structural similarity between each two difference point cloud subsets is obtained based on the distance distribution between the point cloud data matched by the different difference point cloud subsets between each two single buildings in the basic building group, and specifically includes:

[0025] Any two different difference point cloud subsets between any two single buildings in the basic building group are taken as a first difference subset and a second difference subset, respectively.

[0026] ICP matching is performed on the first difference subset and the second difference subset to obtain matching point clouds of each difference point cloud subset corresponding to the other difference point cloud subset.

[0027] For any one of the first difference subset and the second difference subset, the minimum Euclidean distance between each point cloud data in the difference point cloud subset and the coincident point cloud set is taken as the adhesion distance of each point cloud data, and the adhesion matching degree of the difference point cloud subset is determined based on the proportion of the matching successful point cloud data in the difference point cloud subset and the difference between the adhesion distance of each point cloud data in the difference point cloud subset and the adhesion distance of each point cloud data in the other difference point cloud subset.

[0028] The minimum value of the adhesion matching degree in the first difference subset and the second difference subset is taken as the structural matching degree between the first difference subset and the second difference subset.

[0029] Preferably, the base necessity between any two single buildings in the base building group is obtained according to the number of difference point cloud subsets in each common structural class cluster between each two single buildings in the base building group, and specifically comprises:

[0030] For any two single buildings in the base building group, the number of difference point cloud subsets in each common structural class cluster is normalized to obtain a weight coefficient of each common structural class cluster, the number of difference point cloud subsets in each common structural class cluster is weighted and summed using the weight coefficient, and the base necessity between any two single buildings in the base building group is obtained in combination with the number of common structural class clusters.

[0031] Preferably, the modeling scheme of the shared base BIM model is formulated according to each building group in each base building group, and specifically comprises:

[0032] For any building group in any base building group, the product of the negative correlation coefficient of the number of single buildings in the building group and the negative correlation coefficient of the mean value of the base necessity between each single building in the building group and each single building in the base building group where the building group is located is taken as the necessity index of the building group.

[0033] When the necessity index of the building group in the base building group is greater than or equal to a preset fine threshold, each single building in the building group is modeled separately; when the necessity index of the building group in the base building group is less than the preset fine threshold, all single buildings in the building group are modeled together; wherein the overlapping point cloud data of all single buildings in the same base building group constitutes a shared base building model of the base building group.

[0034] Preferably, the base building group is divided into each building group using the base necessity, and specifically comprises:

[0035] Using the DBSCAN clustering algorithm, all single buildings in the base building group are clustered based on the base necessity between different single buildings in each base building group, to obtain each building group in the base building group.

[0036] Preferably, each base building group is obtained by dividing the single buildings using the comprehensive effectiveness, and specifically comprises:

[0037] By using the K-means clustering algorithm, all single buildings are clustered to obtain each basic building group based on the comprehensive effectiveness between different single buildings, and the comprehensive effectiveness between all single buildings in each basic building group is greater than the preset effective threshold.

[0038] The embodiment of the application has at least the following beneficial effects:

[0039] After collecting point cloud data, the embodiment of the application performs feature analysis on each two single buildings, judges whether there is a basic structural loss between the single buildings by using the matching relationship of the difference part and the coincidence part, evaluates the comprehensive effectiveness of the shared building model of the two single buildings, and further divides the single buildings with greater sharing possibility into the same group. Then, by analyzing the structural and adhesion mode similarity of the secondary modeling components represented by the difference part, the basic necessity of the common structure part is evaluated, and further, by dividing the single buildings in the same group into small groups, different ways are used to construct the BIM model of different groups, so that the accuracy of the model construction is ensured while the efficiency of the model construction is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art and the advantages thereof, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can be obtained from these drawings without creative labor.

[0041] Figure 1 is a step flow chart of a building group model simulation construction method provided by the application based on GIS+BIM;

[0042] Figure 2 is a step flow chart of a method for obtaining the comprehensive effectiveness between each two single buildings provided by the application;

[0043] Figure 3 is a step flow chart of a method for obtaining the basic necessity provided by the application. DETAILED DESCRIPTION

[0044] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined object of the application, the following describes in detail the specific implementation, structure, features and effects of a building group model simulation construction method based on GIS+BIM according to the present application, with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0046] The specific scheme of the building group model simulation construction method based on GIS+BIM provided by the present application is described below in detail with reference to the accompanying drawings.

[0047] The main purpose of the present application is that there are often a large number of structures similar or layout similar buildings in a building group, and the existing method cannot effectively and normatively utilize these similarities, resulting in low modeling efficiency. In the present application, by grouping single buildings, the BIM model is constructed in different ways for different groups in the form of basic building model and secondary modeling, so as to ensure the accuracy of model construction while ensuring the efficiency of model construction.

[0048] Please refer to Figure 1 which shows the step flow chart of the building group model simulation construction method based on GIS+BIM provided by one embodiment of the present application, which comprises the following steps:

[0049] Step S100, obtaining point cloud data of each single building.

[0050] The three-dimensional point cloud data of the building group is obtained by using unmanned aerial vehicle laser scanning. After collecting the data, the point cloud data is operated to remove measurement noise and outliers. Ground filtering is applied to separate the ground and building points to remove irrelevant ground points, so as to obtain single buildings. CSF cloth simulation filtering can be used for filtering processing, which is a known technology and will not be described in detail here.

[0051] It should be understood that the building group includes several single buildings, and each single building includes several point cloud data for the construction of the BIM model. Each single building has spatial independence, for example, for a building group of a residential complex, the single building can be a single residential building.

[0052] Since the number of building units of the building group is huge, the commonality of the basic building model of the building unit needs to be considered, and the precision of the model and the efficiency of the BIM model construction also need to be balanced. Therefore, step S200 aims to improve part of the modeling efficiency, by analyzing the similar features of the main structures between the single buildings, the single buildings are grouped, so that the single buildings in the same group can reuse the same basic building structure, and steps S300 and S400 aim to ensure the fine degree of modeling, and the difference between the single buildings is modeled on the basis of the basic building structure, that is, the modeling accuracy of the BIM model is ensured on the basis of improving the modeling efficiency.

[0053] In step S200, according to the coincidence part and the movement matching condition between the difference part of the point cloud data between each two single buildings, and combining the point cloud distribution of the difference part, the comprehensive effectiveness between each two single buildings is obtained; the single buildings are divided by using the comprehensive effectiveness to obtain each basic building group.

[0054] Considering the large number of single buildings of the building group, in order to make the single buildings in the same group share the same basic building structure, it is necessary to analyze the coincidence degree of the point cloud data between different single buildings, so that the basic building model can meet the consistency of the basic building structure. Based on this, this step mainly includes two aspects, the first aspect evaluates the possibility that each two single buildings can use the common basic building structure for modeling, and the second aspect divides the groups for all single buildings.

[0055] As a specific example, as shown in Figure 2 The method for obtaining the comprehensive effectiveness between each two single buildings can be realized by steps S201 to S204.

[0056] In step S201, any two different single buildings are taken as a first building and a second building respectively; the two single buildings are matched to obtain the coincident point cloud set and the non-coincident point cloud set between the first building and the second building.

[0057] The non-coincident point cloud set includes several difference point cloud subsets of the first building or the second building.

[0058] More specifically, the embodiment utilizes the ICP point cloud registration algorithm to register any two single buildings. In the registration logic of the ICP algorithm, the matching process is a one-way process of finding the optimal solution. For example, in the matching result of the first building and the second building, the point cloud data in the first building can find the optimal matching point cloud in the second building, and the point cloud data in the second building can find the optimal matching point cloud in the first building. The overlapping part between the optimal matching point clouds corresponding to the first building and the second building is taken as the overlapping point cloud set between the first building and the second building.

[0059] Further, all point cloud data in the first building except the overlapping point cloud set constitutes the non-overlapping point cloud set of the first building, and all point cloud data in the second building except the overlapping point cloud set constitutes the non-overlapping point cloud set of the second building. Using the DBSCAN clustering algorithm, based on the Euclidean distance between the point cloud data, all point cloud data in the non-overlapping point cloud set is clustered to obtain several difference point cloud subsets in each non-overlapping point cloud set, that is, one clustering cluster in the clustering result corresponds to one difference point cloud subset.

[0060] At this point, for any two different single buildings, an overlapping point cloud set and a non-overlapping point cloud set corresponding to each single building can be obtained. The overlapping point cloud set represents the same structure part between the two single buildings, which can theoretically be used as the basic building structure of the two single buildings. The non-overlapping point cloud set represents the difference structure part of one single building from the other single building. Dividing the non-overlapping point cloud set into several subsets provides a data basis for subsequent analysis of the similarity between the local structures corresponding to each subset.

[0061] Step S202, obtaining the basic inhibitory index of each difference point cloud subset according to the data matching between each difference point cloud subset and the overlapping point cloud set.

[0062] In order to determine whether the non-overlapping phenomenon of the difference point cloud subset corresponding to the two single buildings is caused by spatial position offset or geometric morphology difference, a moving matching analysis is performed on each difference point cloud subset and the overlapping point cloud data to eliminate the difference in spatial position and analyze the real geometric morphology difference. This determination result is directly related to the determination result of whether the basic building structure of the two single buildings can be shared. For example, if the non-overlapping phenomenon of the difference point cloud subset between the two single buildings is caused by position offset, it means that the difference point cloud subset is a main structure missing part, which will inhibit the sharing result of the basic building structure. If it is geometric morphology difference, it means that the non-overlapping phenomenon of the difference point cloud subset is caused by ordinary component difference phenomenon, which does not affect the sharing result of the basic building structure.

[0063] Based on this feature, the feature of each difference point cloud subset corresponding to any two different single buildings is analyzed, and the suppression degree of each difference point cloud set relative to the common basic building structure of the two single buildings is judged.

[0064] In the first step, the direction from the center point cloud of each difference point cloud subset to the center point cloud of the coincident point cloud set is taken as the moving matching direction of each difference point cloud subset.

[0065] In this embodiment, the centroid of all point cloud data contained in each difference point cloud subset in the spatial range is taken as the center point cloud, and the centroid of all point cloud data in the spatial range in the coincident point cloud set is taken as the center point cloud. In other embodiments, the implementer can select according to the specific implementation scene. The center point cloud represents the point cloud data in the center of the point cloud data set.

[0066] In the second step, each difference point cloud subset is moved according to the preset moving step and the moving matching direction.

[0067] In this embodiment, a corresponding preset moving step is set for each difference point cloud subset. Specifically, the Euclidean distance between the center point cloud of each difference point cloud subset and the center point cloud of the coincident point cloud set is taken as the preset moving step corresponding to each difference point cloud subset. In other embodiments, the implementer can set it according to the specific implementation scene. Along the moving matching direction corresponding to each difference point cloud subset, all point clouds in each difference point cloud subset are moved synchronously according to the moving step, so that the geometric shape of the difference point cloud subset itself is unchanged during the moving process.

[0068] It should be noted that the moving stop condition in this embodiment is that the total moving step is equal to the Euclidean distance between the two center point clouds when not moving, and the moving is stopped.

[0069] In the third step, after each moving, the difference point cloud subset and the coincident point cloud set are matched by using a matching algorithm, and the proportion of the point cloud data matched successfully after each moving of the difference point cloud subset is taken as the basic matching degree of each difference point cloud subset after each moving.

[0070] For any difference point cloud subset, after each moving, the ICP algorithm is used to match the moved difference point cloud subset and the coincident point cloud set, that is, the point cloud data with the optimal matching with the difference point cloud subset is obtained in the coincident point cloud set, that is, the point cloud data matched successfully in the difference point cloud subset can find the corresponding optimal matching point cloud data in the coincident point cloud set.

[0071] ​Further, the ratio between the matching successful point cloud data in each difference point cloud subset after each movement and the total number of all point cloud data in the difference point cloud subset is taken as the basic matching degree of each difference point cloud subset after each movement, which reflects the matching degree between the difference point cloud subset and the coincident point cloud set after each movement.

[0072] So far, for each difference point cloud subset, a matching evaluation result between the difference point cloud subset and the coincident point cloud set after each movement can be obtained, which reflects the evaluation result of the movement matching between each difference part and the coincident part of the two single buildings. That is, the movement matching process does not change the geometric consistency evaluation result of the difference point cloud subset and the coincident point cloud set, but changes the spatial position consistency evaluation result between the two, so that the movement matching result can more accurately reflect the real geometric matching relationship.

[0073] In the process of translating each difference part, if the difference part and the coincident part have a continuously high matching degree, it means that the difference part and the coincident part are highly consistent in set form and scale, and only the spatial position is offset. This is due to the missing of local main structures such as floors in the point cloud data, which leads to the difference part not being identified as the coincident part. When the local main structure is missing, the difference structure of the missing part has a certain inhibitory effect on the implementability of the two single buildings sharing the same basic building structure, affecting the effect of the basic building model construction.

[0074] Fourthly, based on the variance and mean of the basic matching degree of each difference point cloud subset corresponding to all movements, the basic inhibitory index of each difference point cloud subset is determined.

[0075] If the difference point cloud subset is actually the missing part of the main structure corresponding to the coincident point cloud (for example, the first building is one floor less, and the difference point cloud subset is the point cloud of the missing floor), at this time, the point cloud geometry of the two is completely consistent, but due to the spatial position offset (the missing floor is above / below the coincident point cloud), the degree of initial matching of the two single buildings is low (that is, the difference point cloud subset and the coincident point cloud set are not coincident).

[0076] In the movement matching process of the difference point cloud subset, each movement will reduce the position deviation of the two groups of point clouds. When the missing floor is moved to the position where it should be, the spatial positions of the two groups of point clouds are completely aligned, and the ICP matching degree will significantly increase, and the matching degree will continue to remain high in the subsequent movement, forming a data feature performance of the difference point cloud subset in the movement process, that is, a high mean and low fluctuation of the basic matching degree. In this case, the effect of movement matching is to make the point clouds that are misjudged as not matching due to position offset show the real matching result of geometric consistency, which provides the basis for calculating the basic inhibitory.

[0077] If the difference point cloud subset is a component that is completely different in geometric shape from the coincident point cloud set (for example, the coincident point cloud set is a wall, and the difference point cloud subset is a balcony), there is an essential difference in the geometric shape of the two groups of point clouds (for example, the relative distances of the points in the point clouds and the structural contour are completely different).

[0078] Even if the spatial position deviation is eliminated through the movement matching operation, the ICP matching algorithm will always have a low matching degree due to the geometric shape mismatch. The basic matching degree generated by the movement of a difference point cloud subset will have a low mean value and a high fluctuation. That is, the matching degree fluctuates greatly when the difference point cloud subset moves to different positions, but the overall matching degree is still low. In this case, the effect of movement matching is to further verify the conclusion that the geometric shapes of the two groups of point clouds are inconsistent, to avoid misjudgment due to position deviation, and to ensure the accuracy of the basic inhibitory calculation.

[0079] As a specific example, for any difference point cloud subset, the product of the mean value and the negative correlation coefficient of the variance of the basic matching degree corresponding to all the movements of the difference point cloud subset is taken as the basic inhibitory index of the difference point cloud subset.

[0080] More specifically, taking the a-th difference point cloud subset as an example, the method for obtaining the basic inhibitory index of the a-th difference point cloud subset can be expressed by the following formula: ; wherein, represents the basic inhibitory index of the a-th difference point cloud subset corresponding to the monomer building A and the monomer building B, represents the mean value of the basic matching degree corresponding to all the movements of the a-th difference point cloud subset, represents the variance of the basic matching degree corresponding to all the movements of the a-th difference point cloud subset, represents an exponential function with the natural constant e as the base.

[0081] The greater the value, the higher the matching degree of the a-th difference point cloud subset with the coincident point cloud set during the movement, and the more consistent the point cloud data in the a-th difference point cloud subset with the point cloud data in the coincident point cloud set in terms of geometric shape. At the same time, the greater the value of the negative correlation coefficient of the variance, the greater the value, the more stable the matching degree of the two during the movement, which is consistent with the characteristics of the missing main structure. At this time, the greater the value of the corresponding basic inhibitory index, the more the a-th difference point cloud subset belongs to the missing main structure component, which will have a greater inhibitory effect on the implementability of one basic building structure of the monomer building A and the monomer building B, and will affect the accuracy of the basic modeling.

[0082] Step S203, obtaining the failure characteristic factor of each single building according to the number of point cloud data contained in each difference point cloud subset corresponding to each single building and the basic inhibitory index.

[0083] Since the absence of the main part often leads to the point cloud data belonging to the missing part being unable to obtain matching points when performing ICP matching, if the difference point cloud subset belongs to the missing part of the main structure, a large amount of point cloud data is contained in the difference point cloud subset. For example, assuming that a difference point cloud subset is a small part, such as a balcony, a bay window, etc., the number of point clouds contained in the corresponding difference point cloud subset is small, and such a difference does not affect the main structure of the building, and has less interference with the common analysis of the basic model. If a difference point cloud subset is a large part of the missing main structure, for example, the first building is one floor less than the second building, the number of point clouds contained in the corresponding difference point cloud subset is large, and such a difference directly destroys the consistency of the main structure, which is the core obstacle of the common analysis of the basic model.

[0084] Based on this, in the process of quantifying the failure characteristic factor of each single building, the number and size of the point cloud data of the difference point cloud subset and the inhibitory degree are comprehensively evaluated, and the failure degree of the strategy of constructing a BIM model for each single building by the way of "common basic model + secondary modeling" is more accurately represented.

[0085] Specifically, for any single building in the first building or the second building, the failure characteristic factor of the single building is determined based on the product of the normalized result of the number of all point cloud data in each difference point cloud subset and the basic inhibitory index.

[0086] More specifically, the failure characteristic factors of the first building and the second building are obtained in the same way, and in this embodiment, the first building is taken as an example for description, and the calculation formula of the failure characteristic factor of the first building can be represented as:

[0087] ; wherein, represents the failure characteristic factor of the first building, represents the total number of difference point cloud subsets of the first building relative to the second building, represents the number of point cloud data contained in the xth difference point cloud subset of the first building relative to the second building, represents the basic inhibitory index of the xth difference point cloud subset of the first building relative to the second building. is a normalization function, for example, maximum and minimum normalization can be used, which is not limited here.

[0088] The failure characteristic factor represents the risk degree shared by the base models of the first building and the second building, and reflects the similarity of the main structures of the first building and the second building. The specific consequence of high failure is that the base model constructed based on the coincident point cloud set of the two monomer buildings cannot cover the main structure of the building.

[0089] In step S204, the maximum value of the failure characteristic factors of the first building and the second building is negatively correlated and normalized to obtain the comprehensive effectiveness between the first building and the second building.

[0090] The comprehensive effectiveness is used to evaluate the feasibility of sharing the base model of the two monomer buildings. The lower the value, the higher the base failure of the two monomer buildings, which means that the similarity of their main structures is poor, and if the base model is forced to be shared, the BIM modeling data will be distorted.

[0091] For the first building and the second building, the failure characteristic factor represents the base failure of the monomer building, that is, the failure degree of the strategy of sharing the base building structure. Each monomer building corresponds to a value of the failure characteristic factor between the two monomer buildings, and then it is necessary to ensure that the base timeliness of the first building and the second building is relatively small, so as to ensure that there is a greater structural similarity between the first building and the second building, that is, the possibility of sharing the base building structure is higher.

[0092] Therefore, the greater the value of the failure characteristic factor corresponding to the monomer building, the smaller the feasibility and effectiveness of sharing the base building structure between the two monomer buildings. Therefore, the minimum effectiveness between the two monomer buildings is used as the evaluation result of the comprehensive effectiveness between the two monomer buildings.

[0093] As a specific example, the method for obtaining the comprehensive effectiveness between the first building and the second building can be represented by the formula: wherein G represents the comprehensive effectiveness between the first building and the second building, represents the failure characteristic factor of the first building, and represents the failure characteristic factor of the second building, represents the maximum value function, represents the normalization function.

[0094] It should be noted that in this embodiment, the maximum value of the failure characteristic factor between every two different monomer buildings is minimized and normalized, and then the comprehensive effectiveness between every two different monomer buildings can be obtained in the form of the above formula. The greater the value of the comprehensive effectiveness, the higher the possibility of sharing the same base building structure between the two different monomer buildings.

[0095] Finally, the comprehensive effectiveness is used to divide all different single buildings into groups, and the structural similarity between each single building in each group is high, and the feasibility of sharing the same basic building structure is high.

[0096] Specifically, the K-means clustering algorithm is used to cluster all single buildings based on the comprehensive effectiveness between different single buildings, and each basic building group is obtained, and the comprehensive effectiveness between all single buildings in each basic building group is greater than the preset effective threshold.

[0097] More specifically, in the first step, all single buildings are clustered according to the comprehensive effectiveness between each two different single buildings, and the value of the number of clustering clusters K is 2, that is, two clustering clusters are obtained in the first step, and it is verified whether the comprehensive effectiveness between each two single buildings in the clustering cluster meets the requirement of the effective threshold.

[0098] In the second step, the clustering cluster corresponding to the comprehensive effectiveness between each two single buildings in the clustering cluster greater than the preset effective threshold is taken as a basic building group. For the clustering cluster that does not meet the threshold requirement, the number of clustering clusters K is kept as 2, the K-means clustering is re-executed for the clustering cluster, and the judgment is performed according to the step.

[0099] In the third step, if neither of the two clustering clusters meets the condition, that is, the comprehensive effectiveness between the single buildings in the two clustering clusters is less than or equal to the effective threshold, the number of clustering clusters is increased by one, the clustering and verification operations are re-executed, and the judgment method is the same as the foregoing, until all single buildings are divided into different basic building groups, which will not be described in detail. In this embodiment, the value of the effective threshold is 0.8, and the implementer can set it according to the specific implementation scene.

[0100] It should be noted that the comprehensive effectiveness represents the similarity between each two single buildings, and when the clustering operation is performed, the comprehensive effectiveness can be processed by negative correlation to serve as the clustering distance between each two single buildings, and the method of negative correlation processing is a known technology, which is not limited herein.

[0101] In step S300, the structural similarity of the difference part of the point cloud data is analyzed according to the distance distribution between the difference part of the point cloud data between each two single buildings in the basic building group, and the basic necessity between each two single buildings in the basic building group is obtained.

[0102] The single buildings in the same basic building group can be modeled using the same building base model, but in the secondary modeling process, there is a case of reuse of secondary modeling structure components, so in the basic building group, the single buildings that can reuse structure components are divided into the same group, and then in the modeling process, the reusability of the structure components in the secondary modeling process is improved.

[0103] Because in the secondary modeling process, the secondary modeling parts of the same type of building have reusability, for example, the main structures of the same type of building are the same, but there are balcony structures in some positions, and because there is structural similarity between the balcony structures, there is structural reusability of small components when constructing the BIM model of the balcony structure, so in order to evaluate the reusability, the structural degree of difference between different single buildings needs to be determined.

[0104] As a specific example, as shown in Figure 3 The basic necessity acquisition method can be implemented by steps S301 to S303.

[0105] In step S301, the structural degree between each two difference point cloud subsets is obtained according to the distance distribution of the matching point cloud data between each two difference point cloud subsets in the basic building group.

[0106] First, any two different difference point cloud subsets between any two single buildings in the basic building group are taken as a first difference subset and a second difference subset.

[0107] In this embodiment, for convenience of description, any two different single buildings in any one basic building group are respectively denoted as a first feature building and a second feature building, and then any two different difference point cloud subsets between the first feature building and the second feature building are respectively taken as a first difference subset and a second difference subset.

[0108] Second, ICP matching is performed on the first difference subset and the second difference subset, and matching point clouds of the point cloud data of each difference point cloud subset in the other difference point cloud subset are respectively obtained.

[0109] It should be noted that the ICP matching algorithm is a known technology, and will not be described in detail here. At this point, for the matching point cloud data in the matching result, the point cloud data of the first difference subset in the second difference subset (the other difference point cloud subset) corresponding to the matching point cloud data can be determined and denoted as the matching point cloud. Similarly, for the matching point cloud data in the matching result, the point cloud data of the second difference subset in the first difference subset (the other difference point cloud subset) corresponding to the matching point cloud data can be determined and denoted as the matching point cloud.

[0110] Thirdly, for any one of the first difference point cloud subset and the second difference point cloud subset, the adhesion matching degree of the difference point cloud subset is evaluated.

[0111] In the embodiment, the adhesion matching degree of the first difference point cloud subset and the second difference point cloud subset is obtained in the same way, and thus the embodiment is described by taking any one of the first difference point cloud subset and the second difference point cloud subset as an example.

[0112] Specifically, the minimum Euclidean distance between each point cloud data in the difference point cloud subset and the coincident point cloud set is taken as the adhesion distance of each point cloud data, and the adhesion matching degree of the difference point cloud subset is determined based on the proportion of the matching successful point cloud data in the difference point cloud subset and the difference between the adhesion distance of each point cloud data in the difference point cloud subset and the adhesion distance of each point cloud data in the other difference point cloud subset.

[0113] More specifically, the first difference point cloud subset is taken as an example, and a Euclidean distance can be calculated between any one of the point cloud data in the first difference point cloud subset and each point cloud data in the coincident point cloud set, and then the minimum Euclidean distance is taken as the adhesion distance of any one of the point cloud data in the first difference point cloud subset.

[0114] The first difference point cloud subset represents the structural difference part between the first feature building and the second feature building, and can represent a component for secondary modeling; the coincident point cloud set represents the structural coincident part between the first feature building and the second feature building, and can represent a main structure for basic modeling; and thus the adhesion distance of each point cloud in the first difference point cloud subset can represent the connection quantitative distance between the secondary modeling component and the main structure.

[0115] Further, as a specific example, the adhesion matching degree of the first difference point cloud subset can be represented by the following formula: wherein, represents the adhesion matching degree of the first difference point cloud subset, represents the basic matching degree of the first difference point cloud subset after ICP matching, and M represents the total number of the matching successful point cloud data in the first difference point cloud subset, represents the adhesion distance of the mth matching successful point cloud data in the first difference point cloud subset, the adhesion distance of the matching point cloud corresponding to the mth matching successful point cloud data in the first difference point cloud subset in the second difference point cloud subset, represents the exponential function with the natural constant e as the base.

[0116] It should be noted that the method for obtaining the basic matching degree corresponding to the first difference subset is similar to the method for obtaining the basic matching degree of the difference point cloud subset in step S202. That is, ICP matching is performed on the first difference subset and the second difference subset, and the ratio between the number of successfully matched point cloud data in the first difference subset and the total number of point cloud data in the first difference subset is used as the basic matching degree corresponding to the first difference subset.

[0117] Basic matching degree It reflects the proportion of point cloud data that were successfully matched in the first difference subset, and thus reflects the degree of similarity in geometric shape between the first difference subset and the second difference subset. In other words, the larger the amount of successfully matched point cloud data, the greater the structural similarity between the two difference point cloud subsets.

[0118] The negative correlation coefficient represents the difference in the corresponding subject connectivity distance between the point cloud data in the first difference subset and the matched point cloud in the second difference subset. It reflects the similarity of the metric distance between the first and second difference subsets of the secondary modeling components. The larger the value, the higher the degree of matching between the first and second difference subsets using the same secondary modeling components.

[0119] Thus, the adhesion matching degree of the first difference subset comprehensively evaluates the matching between the first difference subset and the second difference subset from two aspects: geometric morphological similarity and connection metric distance similarity.

[0120] The fourth step is to take the minimum adhesion matching degree between the first and second difference subsets as the degree of homostructure between the first and second difference subsets.

[0121] Specifically, a higher degree of adhesion matching between the first and second difference subsets indicates a greater likelihood that they belong to the same modeled component. Therefore, the minimum value between the two is used for measurement. The degree of structural similarity between the first and second difference subsets characterizes the probability that they belong to the same structural component. The larger the value, the greater the similarity of the point cloud data between the first and second difference subsets, and the greater the probability that they belong to the same structural component.

[0122] Step S302: Using the degree of similarity in structure, classify all the difference point cloud subsets between every two individual buildings in the basic building group to obtain several common structure clusters between every two individual buildings.

[0123] Specifically, for all difference point cloud subsets between any two single buildings in the basic building group, the DBSCAN clustering algorithm is used to cluster the structural similarity of each two difference point cloud subsets to obtain a plurality of common structural clusters. It should be understood that the structural similarity between each two difference point cloud subsets can be used as a clustering measurement distance by performing negative correlation processing on the structural similarity in the clustering process, and details are not described herein.

[0124] In step S303, the basic necessity between each two single buildings in the basic building group is obtained according to the number of difference point cloud subsets in each common structural cluster between each two single buildings in the basic building group.

[0125] For the first feature building and the second feature building, each common structural cluster represents a secondary modeling component with the same building structure composed of a difference point cloud subset. When evaluating whether it is necessary to use the basic model and secondary modeling to construct the BIM model between two single buildings in the same building group, in order to balance the efficiency and accuracy of BIM modeling, it is necessary to ensure that the number of structural components of the first feature building and the second feature building for secondary modeling is as small as possible, that is, the fewer common structural clusters, the more the difference between the first feature building and the second feature building is mostly the same building structure, and there is no need to spend a lot of time and effort to model each difference part.

[0126] Therefore, the basic necessity is evaluated based on the number of common structural clusters. Specifically, for any two single buildings in the basic building group, the number of difference point cloud subsets in each common structural cluster is normalized to obtain a weight coefficient of each common structural cluster, and the number of difference point cloud subsets in each common structural cluster is weighted and summed using the weight coefficient, and the number of common structural clusters is combined to obtain the basic necessity between any two single buildings in the basic building group.

[0127] More specifically, the first feature building and the second feature building are still used as an example for illustration, and the calculation formula of the basic necessity between the first feature building and the second feature building can be represented as:

[0128]

[0129] wherein, represents the basic necessity between the first feature building and the second feature building, represents the total number of common structural clusters corresponding to the first feature building and the second feature building, represents the number of difference point cloud subsets in the cth common structure cluster corresponding to the first feature building and the second feature building, is a normalization function, represents the weight coefficient of the common structure cluster. is a normalization function, which is a max-min normalization function in this embodiment.

[0130] It should be noted that the number of difference point cloud subsets between each two monomer buildings can be calculated and obtained , and then processed using the max-min normalization method. The greater the value of is, the greater the proportion of the same building structure in the difference part of the first feature building and the second feature building, and The greater the value of is, the fewer the building structures in the difference part, and the fewer the building structures required for secondary modeling.

[0131] Through the weighted processing of the number of difference point cloud subsets contained in the common structure cluster, the large-scale common structure cluster occupies a dominant position in the overall reuse value summation, accurately filters the interference of small and many low-value clusters, not only considers how many reusable difference point cloud subsets each cluster contains, but also highlights the dominant role of large-scale clusters on reuse value, and finally supports the accuracy of basic necessity. The greater the value of the basic necessity, the higher the necessity of the secondary modeling strategy for the first feature building and the second feature building, and the higher the reusability of the secondary modeling component.

[0132] Step S400, divide the basic building large group into each building small group using the basic necessity; according to each building small group in each basic building large group, develop a modeling scheme for sharing a basic BIM model, and combine GIS to simulate a building group model.

[0133] First, monomer buildings with similar reusability of secondary modeling components in the same basic building large group are divided into the same building small group.

[0134] Specifically, using the DBSCAN clustering algorithm, based on the basic necessity between different monomer buildings in each basic building large group, all monomer buildings in the basic building large group are clustered to obtain each building small group in the basic building large group.

[0135] It should be noted that, by analogy with the clustering process of the common structure cluster, the basic necessity between each two monomer buildings represents the similarity, feasibility of secondary modeling and reusability of secondary modeling components of the two monomer buildings. In the clustering process, the clustering operation can be performed by negatively correlating the basic necessity between each two monomer buildings to obtain the clustering measurement distance, which will not be described in detail here.

[0136] Each building group contains many repetitive secondary modeling structural components for individual buildings. However, the secondary modeling structural components of some individual buildings are different from those of most individual buildings. In order to ensure the accuracy and effectiveness of BIM modeling, it is necessary to accurately model these secondary modeling structural components. These secondary modeling structural components exist in building groups with a small number of individual buildings and low foundation necessity between individual buildings.

[0137] Based on this characteristic, considering the number distribution of individual buildings and the necessity of foundations within each basic building group, we can further analyze whether to perform secondary modeling of the same structure for the building group or to use more refined individual modeling to formulate a personalized modeling scheme for a shared basic BIM model. This improves modeling efficiency while ensuring modeling accuracy.

[0138] Then, the number of individual buildings and the necessity of the foundation are analyzed for each building subgroup within each basic building group, and the necessity index of each building subgroup is evaluated.

[0139] Specifically, for any building subgroup within any basic building group, the product of the negative correlation coefficient of the number of individual buildings in the building subgroup and the negative correlation coefficient of the mean basic necessity between each individual building in the building subgroup and each individual building in the basic building group to which the building subgroup belongs is used as the necessity index of the building subgroup.

[0140] More specifically, the number of individual buildings included in a building group. negative correlation coefficient As the first coefficient of the building group; each individual building in the building group corresponds to a foundation necessity value with different individual buildings in the foundation building group to which the building group belongs, and the mean of all foundation necessities corresponding to the building group is calculated. negative correlation coefficient The second coefficient is obtained. Finally, the product of the first and second coefficients is the necessity index for this building group.

[0141] Number of individual buildings included in the architectural group This reflects the reuse cost coefficient allocated to each building. A higher value indicates a higher reuse cost per building and lower returns, thus suggesting greater feasibility for detailed modeling and a higher necessity index value. The lower the fundamental necessity between the building group and its parent building group, the higher the value of the second coefficient, indicating a higher risk of inaccurate reuse of secondary components and a greater need for precise modeling; thus, a higher necessity index value is required. The necessity index characterizes the degree of necessity for the building group to adopt individual detailed modeling.

[0142] In the embodiment, the product of the first coefficient and the second coefficient corresponding to each building group is also normalized to obtain the necessity index. The normalization method can adopt the maximum-minimization normalization method, which is not limited here.

[0143] Further, after the division of the monomer building models into groups and subgroups, different modeling strategies need to be developed for different groups.

[0144] In the first step, the BIM modeling is performed on the coincident point cloud data of all monomer buildings in the same basic building group to obtain the common basic building model of each basic building group. That is, the coincident point cloud data of all monomer buildings in the same basic building group constitutes the common basic building model of the basic building group.

[0145] In the second step, for any building group in the basic building group, when the necessity index of the building group in the basic building group is greater than or equal to the preset fine threshold, the individual modeling is performed on each monomer building in the building group.

[0146] When the necessity index of the building group is greater than or equal to the fine threshold, it indicates that the number of monomer buildings in the building group is small and the similarity is low, and the reuse benefit of the secondary modeling component of the group is low and the risk is high. Therefore, the secondary precise BIM modeling is performed on the non-coincident distribution based on the building group, which can ensure the accuracy of the modeling result.

[0147] In the third step, when the necessity index of the building group in the basic building group is less than the preset fine threshold, the common modeling is performed on all monomer buildings in the building group.

[0148] When the necessity index of the building group is less than the fine threshold, it indicates that the number of monomer buildings in the building group is large and the similarity is high, and the reuse benefit of the secondary modeling component of the group is high and the risk is low. Therefore, the common structural cluster in the building group is reused as the secondary component model for the secondary construction operation of the BIM model, which can improve the modeling efficiency to a certain extent.

[0149] In the embodiment, the value of the fine threshold is 0.6, which can be set according to the specific implementation scene.

[0150] Finally, according to the modeling result of the BIM model, the simulation operation of the building group model is performed in combination with GIS. The simulation process of the building group model based on BIM and GIS is a conventional application in the field, which is only briefly introduced here.

[0151] Obtain the GIS data such as terrain data, road, land boundary, building distribution coordinates in the range of the region to be modeled, align each single point cloud data in the BIM model with the GIS coordinate system, realize the accurate positioning of the building group in the real geographical environment, superimpose the BIM model to the geographical coordinates and terrain environment provided by GIS, form the building group scene corresponding to the real environment, bind the building model with the GIS attribute data, render the whole building group in the three-dimensional scene and carry out simulation application.

[0152] In summary, for the problem of the implementability judgment of the basic model and the secondary modeling for any two single building models, the application uses the matching relationship of the difference distribution and the coincident point cloud set to judge whether there is a basic structural loss, thereby evaluating the possibility of the common use of the basic model; for the problem of the large amount of scattered parts of the secondary modeling part, the application analyzes the similarity of the structure and the adhesion mode between the secondary modeling parts, judges the same structure, and further evaluates the basic necessity of the common structure cluster; in the specific implementation, the single buildings are grouped, different groups are used to construct the BIM model in different ways, thereby ensuring the accuracy of the model construction while ensuring the efficiency of the model construction.

[0153] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A GIS+BIM-based building group model simulation construction method, characterized in that, The method comprises the following steps: Obtaining point cloud data of each single building; According to the coincidence part and the moving matching condition between the difference part of the point cloud data between each two single buildings, and combining the point cloud distribution of the difference part, the comprehensive effectiveness between each two single buildings is obtained; the single buildings are divided by using the comprehensive effectiveness to obtain each basic building group; According to the distance distribution between the point cloud data of the difference part between each two single buildings in the basic building group, the same structure degree of the point cloud data of the difference part is analyzed to obtain the basic necessity between each two single buildings in the basic building group; Using the basic necessity, the basic building group is divided into each building group; according to each building group in each basic building group, a modeling scheme of a shared basic BIM model is formulated, and a building group model simulation is performed in combination with GIS; According to the coincidence part and the moving matching condition between the difference part of the point cloud data between each two single buildings, and combining the point cloud distribution of the difference part, the comprehensive effectiveness between each two single buildings is obtained; the single buildings are divided by using the comprehensive effectiveness to obtain each basic building group; Any two different single buildings are taken as a first building and a second building respectively; the two single buildings are matched to obtain a coincident point cloud set and a non-coincident point cloud set between the first building and the second building; the non-coincident point cloud set includes several difference point cloud subsets of the first building or the second building; According to the data matching condition between each difference point cloud subset and the coincident point cloud set, a basic inhibition index of each difference point cloud subset is obtained; According to the number of point cloud data contained in each difference point cloud subset corresponding to each single building and the basic inhibition index, an invalidity feature factor of each single building is obtained, including: for any one single building in the first building or the second building, the invalidity feature factor of the single building is determined based on the product of the normalization result of the number of all point cloud data in each difference point cloud subset and the basic inhibition index; The maximum value of the invalidity feature factors of the first building and the second building is negatively correlated and normalized to obtain the comprehensive effectiveness between the first building and the second building; According to the data matching condition between each difference point cloud subset and the coincident point cloud set, a basic inhibition index of each difference point cloud subset is obtained; The direction from the center point cloud of each difference point cloud subset to the center point cloud of the coincident point cloud set is taken as the moving matching direction of each difference point cloud subset; According to a preset moving step and the moving matching direction, each difference point cloud subset is moved, and after each movement, the difference point cloud subset and the coincident point cloud set are matched by using a matching algorithm; the proportion of the point cloud data matched successfully after each movement is taken as the basic matching degree of each difference point cloud subset under each movement; Based on the variance and mean value of the basic matching degree corresponding to each difference point cloud subset under all movements, the basic inhibition index of each difference point cloud subset is determined.

2. The GIS+BIM-based building group model simulation construction method according to claim 1, characterized in that, The distance distribution between the point cloud data of the difference part between each two single buildings in the basic building group is analyzed, and the structural similarity of the point cloud data of the difference part is obtained, and the basic necessity between each two single buildings in the basic building group is obtained, specifically including: According to the distance distribution of the matching point cloud data of the different difference point cloud subsets between each two single buildings in the basic building group, the structural similarity between each two difference point cloud subsets is obtained. Using the structural similarity, all difference point cloud subsets between each two single buildings in the basic building group are classified to obtain several common structure clusters between each two single buildings. According to the number of difference point cloud subsets in each common structure cluster between each two single buildings in the basic building group, the basic necessity between each two single buildings in the basic building group is obtained.

3. The GIS+BIM-based building group model simulation construction method according to claim 2, characterized in that, The distance distribution of the matching point cloud data of the different difference point cloud subsets between each two single buildings in the basic building group is analyzed, and the structural similarity between each two difference point cloud subsets is obtained, specifically including: Take any two different difference point cloud subsets between any two single buildings in the basic building group as the first difference subset and the second difference subset respectively. ICP matching is performed on the first difference subset and the second difference subset to obtain the matching point cloud of each difference point cloud subset in the other difference point cloud subset. For any one of the first difference subset and the second difference subset, the minimum Euclidean distance between each point cloud data in the difference point cloud subset and the coincident point cloud set is taken as the adhesion distance of each point cloud data, and based on the proportion of the matching successful point cloud data in the difference point cloud subset and the difference between the adhesion distance of each point cloud data in the difference point cloud subset and the adhesion distance of each point cloud data in the other difference point cloud subset, the adhesion matching degree of the difference point cloud subset is determined. The minimum value of the adhesion matching degree of the first difference subset and the second difference subset is taken as the structural similarity between the first difference subset and the second difference subset.

4. The GIS+BIM-based building group model simulation construction method according to claim 3, characterized in that, The number of difference point cloud subsets in each common structure cluster between each two single buildings in the basic building group is normalized to obtain the weight coefficient of each common structure cluster, and the number of difference point cloud subsets in each common structure cluster is weighted and summed using the weight coefficient, and the number of common structure clusters is combined to obtain the basic necessity between any two single buildings in the basic building group. According to each building group in each basic building group, a modeling scheme of a common basic BIM model is developed, specifically including:

5. The GIS+BIM-based building group model simulation construction method according to claim 1, characterized in that, ​ For any one building group in any one basic building group, the product of the negative correlation coefficient of the number of single buildings in the building group and the negative correlation coefficient of the mean value of the basic necessity between each single building in the building group and each single building in the basic building group where the building group is located is taken as the necessity index of the building group; When the necessity index of the building group in the basic building group is greater than or equal to the preset fine threshold, each single building in the building group is modeled separately; when the necessity index of the building group in the basic building group is less than the preset fine threshold, all single buildings in the building group are modeled together; wherein the overlapping point cloud data of all single buildings in the same basic building group constitutes a common basic building model of the basic building group.

6. The GIS+BIM-based building group model simulation construction method according to claim 1, characterized in that, The basic building group is divided into each building group by using the basic necessity, specifically including: Using the DBSCAN clustering algorithm, the basic necessity between different single buildings in each basic building group is used to cluster all single buildings in the basic building group, and each building group in the basic building group is obtained.

7. The GIS+BIM-based building group model simulation construction method according to claim 1, characterized in that, The single building is divided by using the comprehensive effectiveness to obtain each basic building group, specifically including: Using the K-means clustering algorithm, the comprehensive effectiveness between different single buildings is used to cluster all single buildings to obtain each basic building group, and the comprehensive effectiveness between all single buildings in each basic building group is greater than the preset effective threshold.

Images