BIM-based complex road section multi-specialty design checking method

Abstract

The application relates to the technical field of building model data processing, in particular to a complex road section multi-specialty design checking method based on BIM. The method comprises the following steps: analyzing the difference in the number of attributes and the difference in size data between any two components, and combining the similarity of attribute data between any two components to cluster each component in a professional cluster under each professional category with all the remaining components; obtaining a replacement cluster, determining a retention coefficient based on the number of directly connected components of each component, combining the difference in the replacement coefficient between each component in each professional category and all the components in the replacement cluster of the component, and the number of attributes, to obtain a retained component, performing lightweight processing on the BIM model, and checking the BIM model after the lightweight processing. The application accurately and efficiently removes components by considering the professional and attribute differences between similar components, thereby improving the efficiency and accuracy of the BIM model checking of the complex road section.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building model data processing, in particular to a complex road section multi-specialty design checking method based on BIM. BACKGROUND

[0002] As a new type of engineering information management method, building information modeling (BIM) can effectively solve the problems in the design and construction stages and reduce construction costs. However, as the building size becomes larger and the model becomes more complex, the dependence of BIM models on computer hardware and software requirements is also increasing. In the construction of complex road sections such as viaducts and overpasses, the design of bridge BIM models involves multiple specialties and has a large amount of parameter data, which makes the building model processing software such as Revit slow in processing this type of data, thereby affecting the subsequent checking efficiency of multi-specialty design for complex road sections. Therefore, it is necessary to build a lightweight method for BIM models of complex road sections to improve the subsequent checking efficiency of multi-specialty design for BIM models.

[0003] When the prior art performs lightweight processing on BIM models, it usually adopts an instantiation method of directly comparing the similarity of components and retaining one component and removing similar components. However, the prior art does not fully consider the differences in specialties and attributes between similar components, and there may be components that are similar in appearance but different in nature between different specialties, which may lead to incorrect classification of cross-specialty components, thereby removing components that should be retained, resulting in poor component removal effect and damage to the integrity of the final model, thereby reducing the efficiency and accuracy of checking complex road section BIM models. SUMMARY

[0004] To solve the above technical problems, the present application provides a complex road section multi-specialty design checking method based on BIM to solve the existing problems.

[0005] The complex road section multi-specialty design checking method based on BIM of the present application adopts the following technical scheme:

[0006] One embodiment of the present application provides a complex road section multi-specialty design checking method based on BIM, which includes the following steps:

[0007] Obtain the names of all specialty categories in the BIM model of the viaduct project and the attribute data of each component, and perform word segmentation processing on the specialty category names and the attribute data of each component, respectively;

[0008] traverse root nodes of each word in WordNet; count distances from root nodes of all words of each component to root nodes of all words of names of each professional category to determine first distances between each component and each professional category; obtain all synonyms of each word, analyze differences of all synonyms of all words between each component and names of each professional category to determine second distances between each component and each professional category, and form distance vectors with the first distances; cluster all components based on the distance vectors, and filter professional clusters under each professional category based on lengths of distance vectors of cluster centers in each cluster;

[0009] count attribute quantity and size data of each component, analyze differences of attribute quantity and size data between any two components, and determine a substitution coefficient between any two components based on similarity of attribute data between the any two components; cluster each component and all other components in a professional cluster under each professional category based on the substitution coefficient, and filter a substitution cluster of each component under each professional category from all clusters based on a mean value of substitution coefficients between all components in each cluster;

[0010] based on attribute data of all components under each remaining component, to check the BIM model after the lightweight processing.

[0011] Preferably, the first distance between each component and each professional category is a minimum value in distances from root nodes of all words of each component to root nodes of all words of names of each professional category.

[0012] Preferably, the second distance between each component and each professional category is a word shift distance of all synonyms of all words between each component and names of each professional category.

[0013] Preferably, a metric distance in the clustering process of all components based on the distance vectors is an Euclidean distance between the distance vectors.

[0014] Preferably, the professional cluster under each professional category is a cluster with a minimum length of distance vectors of cluster centers in all cluster centers of all cluster groups under each professional category.

[0015] Preferably, the substitution coefficient between any two components is expressed as: wherein, A ij represents the substitution coefficient between the i th component and the j th component; D ij represents similarity of attribute data between the i th component and the j th component; Bij , C ij respectively represent the difference between the i th component and the j th component, the difference of the size data; τ represents a constant greater than 0.

[0016] Preferably, the replacement cluster of each component under each professional category is the cluster with the maximum average replacement coefficient among all clusters of each component under each professional category.

[0017] Preferably, the expression of the retention coefficient of each component under each professional category is: F u,v represents the retention coefficient of component v under professional category u; L v represents the number of components directly connected to component v; N v represents the number of attributes of component v under professional category u; G u,v represents the difference between component v and all components in its replacement cluster under professional category u; ε represents a constant greater than 0.

[0018] Preferably, the remaining component of each component under each professional category is the component corresponding to the maximum retention coefficient in the replacement cluster of each component under each professional category.

[0019] Preferably, the lightweight BIM model based on the attribute data of all components under each remaining component in each professional category is used to check the lightweight processed BIM model, comprising:

[0020] All components under each remaining component in each professional category are denoted as removed components, the attribute data of each removed component is denoted as an attribute set, the attribute set of each removed component is mapped to the remaining component, and all removed components of each remaining component in each professional category are deleted, all professional categories are traversed, and a lightweight processed BIM model is obtained.

[0021] The lightweight processed BIM model is used as the input of the checking software, and the checking result is output.

[0022] The present application has at least the following beneficial effects:

[0023] The application can effectively distinguish components of different professions, avoid cross-professional comparison, and significantly improve the multi-professional design checking efficiency of the complex road section BIM model while ensuring accuracy by using WordNet analysis to calculate the first and second distances and normalize them to form a distance vector, and using a clustering algorithm to divide the components by professional category. Further, the application can identify replaceable clusters within the same profession through multi-professional differentiation and similarity analysis, and select key remaining components based on the structural importance, information value, and representative of the clusters, and use the remaining components to represent other components in the replacement cluster, which not only ensures the integrity of the key information of the model, but also significantly reduces the model data, thereby helping to improve the checking efficiency of the multi-professional BIM model of the complex road section. The application can accurately distinguish multi-professional components through semantic analysis, and calculate similarity and importance based on attributes, sizes, and connection relationships, and select the most representative remaining components in each profession to replace other components in the cluster, which significantly reduces the model data, realizes lightweight BIM model, and ensures the preservation of key information, thereby improving the checking efficiency and accuracy of the multi-professional BIM model of the complex road section. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings needed in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0025] Figure 1 The step flow chart of the BIM-based multi-professional design checking method for complex road sections provided by an embodiment of the present application;

[0026] Figure 2 The schematic diagram of the replacement cluster acquisition process provided by an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the specific embodiments, structure, features and effects of the BIM-based multi-professional design checking method for complex road sections according to the present application are described in detail as follows. 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.

[0028] 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 the present application belongs.

[0029] The specific scheme of the BIM-based complex road section multi-specialty design checking method provided in the application will be specifically described below in combination with the drawings.

[0030] The BIM-based complex road section multi-specialty design checking method provided in an embodiment of the application, in particular, provides the following BIM-based complex road section multi-specialty design checking method, please refer to Figure 1 The method comprises the following steps:

[0031] Step S1: Obtain the names of all specialty categories in the BIM model of the viaduct project and the attribute data of each component, and perform word segmentation processing on the names of each specialty category and the attribute data of each component.

[0032] A viaduct construction project usually involves multiple professional fields, such as civil engineering, surveying and mapping, power, etc. Therefore, before the actual construction of the viaduct project, a BIM three-dimensional model integrating the model construction of each specialty is first constructed, and the BIM model is checked to detect whether hard collision or gap collision will occur between the model components of each specialty model inside the model, and to check whether the attribute parameters of each model construction are reasonable, so as to avoid the problems of construction rework or waste.

[0033] Since the viaduct BIM model has a large amount of parameters, which affects the efficiency of multi-specialty design checking, the BIM model of the viaduct project constructed in this embodiment is subjected to lightweight processing, and the specific process is as follows:

[0034] The BIM model is exported in IFC standard format, and then the exported IFC format file is parsed by an IFC parsing tool such as ifcopenshell, so as to obtain all entities in the IFC file and the attribute data contained by each entity. In this embodiment, the attribute data includes the size data of the entity, i.e., the component, such as length, width, and height, the type, function, position, and material of the component. Different components may contain different types of attribute data.

[0035] Among them, one entity corresponds to one model component in the original BIM model of the viaduct project, and the name of the component is reserved as the Name attribute value of the entity. Since the name of the component may be long, the attribute data of each component obtained is subjected to word segmentation processing, and in this embodiment, the attribute data of each component is divided into multiple words by using the jieba word segmentation tool.

[0036] Among them, the jieba word segmentation tool is a known technology, and the specific process of using it to perform word segmentation on the text will not be described again.

[0037] Further, the major of all designers participating in the BIM model of the viaduct project is obtained manually, and the major category to which each major belongs is obtained according to the major name according to the major table, and finally the professional field set of the BIM model is constructed by the name of the major category. For example: if the major of a designer is road and bridge crossing river engineering, which belongs to the major category of civil engineering, the professional field set is {“civil engineering”, “water conservancy”}. At the same time, similarly, the name of each major category is segmented by using a segmentation tool according to the segmentation processing method of the attribute data of the component, and all the segmented words of the name of each major category are obtained.

[0038] Step S2: traversing the root nodes of each segmented word in WordNet; counting the distance from the root node of all segmented words of each component to the root node of all segmented words of each major category name, determining the first distance between each component and each major category; obtaining all synonyms of each segmented word, analyzing the difference between all synonyms of each component and each major category name, determining the second distance between each component and each major category, and forming a distance vector with the first distance; clustering all components based on the distance vector, and filtering out the professional cluster under each major category based on the length of the distance vector corresponding to the cluster center in each cluster.

[0039] Since the BIM model of the viaduct on the complex section is constructed by integrating model components of multiple professional fields, it contains a large amount of components; and in the same BIM model of the viaduct, there are BIM model components with similar appearance, a large number of repetitions but different spatial positions in each major, such as pile foundation and simply supported beam in the bridge major. Therefore, the model with similar appearance and repeated size in each major can be stored by using the instantiation mechanism, so as to reduce the number of model components and improve the checking efficiency of the BIM model between multiple majors. However, considering that there may be similar components between different majors, the multi-major field can be distinguished according to the characteristic parameters of the model components before comparing the similarity of each model component, and then only the components of the major are compared in the subsequent similarity comparison, avoiding comparing all major components, improving the efficiency and accuracy of component comparison.

[0040] Based on the above analysis, the embodiment traverses the root nodes of each word in the WordNet; the distance between each component and each professional category is determined by counting the distance between the root nodes of all words of each component and the root nodes of all words of the name of each professional category; all synonyms of each word are obtained, the difference between all synonyms of each component and each professional category is analyzed, the second distance between each component and each professional category is determined, and the first distance is combined to form a distance vector; all components are clustered based on the distance vector, and the length of the distance vector corresponding to the cluster center in each cluster is used to filter out professional clusters under each professional category, so as to realize professional classification of different components, specifically:

[0041] In the embodiment, first, the root nodes of each word are traversed in the Chinese WordNet dictionary, wherein the WordNet is essentially a semantic network, in which words are connected by specific relationships, and the hyponym-hypernym relationship is an important relationship, the hypernym is a more general and more general word, and the root node is the most general word at the top of the WordNet network. For example, assuming that there is a component, and the name string of the component is "K10+200 left 3# pile foundation", after word segmentation, [“K10+200”, “at”, “left”, “3#”, “pile foundation”] is obtained, assuming that the hypernym of “pile foundation” is obtained, the WordNet will inform that “pile foundation” is a “foundation”, therefore, “foundation” is the hypernym of “pile foundation”, and the root node is obtained by traversing the entire WordNet dictionary until the hypernym of “pile foundation” cannot be obtained any more. At this time, the most general word of “pile foundation” is obtained, that is, the root node is obtained.

[0042] The Chinese WordNet dictionary is a known technology, and its specific principle will not be described again.

[0043] Further, the first distance between each component and each professional category is determined by counting the distance between the root nodes of all words of each component and the root nodes of all words of the name of each professional category, specifically:

[0044] In the embodiment, the minimum value of the distance between the root nodes of all words of each component and the root nodes of all words of the name of each professional category is taken as the first distance between each component and each professional category.

[0045] It should be noted that the distance between the root nodes in the embodiment is actually the weight of the edge between the root nodes, and the sum of the weights of all edges experienced between the root nodes of each word in each component and the root nodes of each word of the name of each professional category is taken as the distance between the root nodes.

[0046] According to the first distance between each component and each professional category, it can be understood that the first distance is an index for measuring the distance of field belonging relationship between a component and a professional category, and reflects the difference between the field category represented by the component and the field category represented by the professional category. If the first distance between the current component and the current professional category is larger, it means that the difference between the field category to which the current component belongs and the field category of the current professional category is larger, indicating that the possibility that the current component does not belong to the current professional category is larger. Conversely, if the first distance between the current component and the current professional category is smaller, it means that the difference between the field category to which the current component belongs and the field category of the current professional category is smaller, indicating that the possibility that the current component does not belong to the current professional category is smaller, i.e., the possibility that the component belongs to the current professional category is larger.

[0047] Further, the embodiment determines the second distance between each component and each professional category by analyzing the difference of all synonym words of all words between each component and each professional category name, and forms a distance vector with the first distance, specifically as follows:

[0048] As a specific implementation, the embodiment uses synonym word extraction algorithm to obtain all synonym words of each word, and the synonym word extraction algorithm is a known technology, and the specific process of obtaining synonym words will not be described again.

[0049] Further, the embodiment takes the word shift distance of all synonym words of all words between each component and each professional category name as the second distance between each component and each professional category.

[0050] The calculation method of the word shift distance is a known technology, and the specific calculation process will not be described again.

[0051] According to the second distance between each component and each professional category, it can be understood that the second distance reflects the semantic difference between the attribute data of the component and the professional category name, and is also used to represent the possibility that the component belongs to the professional category. If the second distance between the current component and the current professional category is larger, it means that the matching degree between the current component and the current professional category is lower, indicating that the possibility that the current component does not belong to the current professional category is larger. Conversely, if the second distance between the current component and the current professional category is smaller, it means that the matching degree between the current component and the current professional category is higher, indicating that the possibility that the current component does not belong to the current professional category is smaller, i.e., the possibility that the component belongs to the current professional category is larger.

[0052] Further, the embodiment forms a distance vector between each component and each professional category by taking the normalized value of the first distance and the normalized value of the second distance between each component and each professional category.

[0053] It should be noted that, in order to prevent the influence of the data dimension, the first distance and the second distance are normalized in the embodiment, and the maximum and minimum value normalization method is used to normalize the first distance and the second distance in the embodiment. In actual application, as other implementation manners, the implementer can also use other normalization methods such as z-score standardization method according to specific conditions, and the selection of the normalization method is not specially limited in the embodiment.

[0054] The maximum and minimum value normalization method is a known technology, and the specific process of normalizing data by using the method will not be repeated.

[0055] Further, the embodiment clusters all components based on the distance vector, and filters professional clusters under each professional category based on the length of the distance vector corresponding to the cluster center in each cluster. Specifically,

[0056] In the embodiment, all components are used as the input of the clustering algorithm, wherein the Euclidean distance between the distance vectors is set as the measurement distance in the clustering algorithm, the elbow method is used to determine the number of cluster clusters, and all cluster clusters under each professional category are output.

[0057] Further, the cluster cluster with the minimum length of the distance vector corresponding to the cluster center under each professional category is used as the professional cluster under the professional category, which is used to represent the set of components belonging to the same professional field.

[0058] It should be noted that there are many commonly used clustering algorithms, and the k-means clustering algorithm is used to divide all components according to the professional category in the embodiment. In actual application, the implementer can also select other clustering methods such as DPC density peak clustering algorithm according to specific conditions, and the selection of the clustering algorithm is not specially limited in the embodiment.

[0059] The calculation process of the Euclidean distance, the elbow method and the k-means clustering algorithm are all known technologies, and their specific principles and processes will not be repeated.

[0060] So far, the embodiment analyzes the semantic relationship between the component name and the professional category by using WordNet, calculates the first and second distances and normalizes them to form a distance vector, and divides the components according to the professional category by using the clustering algorithm, thereby effectively distinguishing the components of different professions, avoiding cross-professional comparison, and further improving the multi-professional design checking efficiency of the complex road section BIM model while ensuring the accuracy.

[0061] Step S3: count the attribute quantity and size data of each component, analyze the difference in attribute quantity and the difference in size data between any two components respectively, and determine the substitution coefficient between any two components in combination with the similarity of attribute data between any two components, cluster each component in a professional cluster under each professional category with all the remaining components based on the substitution coefficient, and filter out the substitution cluster of each component under each professional category from all clusters based on the mean value of the substitution coefficient between all components in each cluster; count the number of directly connected components of each component, and determine the retention coefficient of each component under each professional category in combination with the difference in substitution coefficient between each component under each professional category and all components in its substitution cluster and the attribute quantity, to obtain the remaining components of each component under each professional category.

[0062] After the components are distinguished by multiple professions, the similarity of the components in the same profession can be measured, so that the components with consistent appearance and inconsistent spatial position are removed. Then, model mapping is performed through only one component to ensure the integrity of the BIM model, thereby reducing the volume of the BIM model while ensuring the integrity of the BIM model, and improving the checking efficiency of the complex road section model.

[0063] Therefore, based on the above analysis, the embodiment counts the attribute quantity and size data of each component, analyzes the difference in attribute quantity and the difference in size data between any two components respectively, and determines the substitution coefficient between any two components in combination with the similarity of attribute data between any two components, clusters each component in a professional cluster under each professional category with all the remaining components based on the substitution coefficient, and filters out the substitution cluster of each component under each professional category from all clusters based on the mean value of the substitution coefficient between all components in each cluster; counts the number of directly connected components of each component, and determines the retention coefficient of each component under each professional category in combination with the difference in substitution coefficient between each component under each professional category and all components in its substitution cluster and the attribute quantity, to obtain the remaining components of each component under each professional category, thereby lightening the BIM model and improving the efficiency and accuracy of the checking of the complex road section BIM model, and the specific process is as follows:

[0064] Since the size attributes of the components with consistent appearance in the same professional field are also consistent, the embodiment first counts the size data of each component, wherein the size data is a vector composed of length, width and height in the size data of the component obtained in step S1, i.e. [length, width, height], and the attribute quantity of each component is counted for subsequent analysis of the similarity between components. The attribute quantity is the type of attribute data in step S1, i.e. the size data of the component, the type of the component, the function, the position and the material. The types of attribute data contained by different components may be different, so the attribute quantities of different components are not the same.

[0065] Further, the embodiment determines the substitution coefficient between any two components by analyzing the difference in attribute quantity and the difference in size data between any two components respectively, and combining the similarity of attribute data between any two components, specifically as follows:

[0066] As a specific implementation, in the embodiment, the expression of the substitution coefficient A ij between the ith component and the jth component is as follows: In the expression, A ij represents the substitution coefficient between the ith component and the jth component; D ij represents the similarity of attribute data between the ith component and the jth component; B ij and C ij respectively represent the difference in attribute quantity and the difference in size data between the ith component and the jth component; τ represents a constant greater than 0, which is preset to prevent the denominator from being 0. In the embodiment, the value of τ is artificially set, and in the embodiment, the value of τ is 0.01. On the premise of ensuring that the denominator is not 0 and not excessively affecting the calculation result, the implementer can also set it according to the specific circumstances, and the embodiment does not make special limitations.

[0067] It should be noted that there are many methods for measuring the similarity between data sets. In the embodiment, all attribute data of each component is combined to form an attribute set, and the Jaccard similarity of the attribute set between the ith component and the jth component is taken as the similarity of attribute data between the ith component and the jth component. In actual application, as other implementation manners, the implementer can also use cosine similarity or the reciprocal of Euclidean distance and other methods for measuring the similarity between data sets according to the specific circumstances. The embodiment does not make special limitations on the selection of the method for measuring the similarity between data sets.

[0068] In addition, it should be understood that since the size data is actually a vector, there are many methods for measuring the difference between vectors. In the embodiment, the DTW distance of the size data between the ith component and the jth component is taken as the difference in size data between the ith component and the jth component. In actual application, the implementer can also use Euclidean distance or Manhattan distance and other methods for measuring the difference between vectors according to the specific circumstances, and the embodiment does not make special limitations.

[0069] In the expression, the calculation method of the DTW distance is a known technology, and the specific calculation process is not described again.

[0070] According to the substitution coefficient between any two components, it can be understood that the substitution coefficient is used to measure the similarity of two components in appearance size, function implementation and name, so as to judge whether they can be substituted for each other; the greater the similarity of attribute data between the ith component and the jth component, the more similar the attributes between the ith component and the jth component, so that the possibility of mutual substitution is greater, and therefore the corresponding substitution coefficient is greater; at the same time, the smaller the difference in the number of attributes between the ith component and the jth component, the smaller the difference in functional attributes between the ith component and the jth component, so that the possibility of mutual substitution is greater, and therefore the corresponding substitution coefficient is greater; in addition, the greater the difference in size data between the ith component and the jth component, the more consistent the sizes of the ith component and the jth component, so that the possibility of mutual substitution is greater, and therefore the corresponding substitution coefficient is greater.

[0071] On the contrary, the smaller the similarity of attribute data between the ith component and the jth component, the less similar the attributes between the ith component and the jth component, so that the possibility of mutual substitution is smaller, and therefore the corresponding substitution coefficient is smaller; at the same time, the greater the difference in the number of attributes between the ith component and the jth component, the greater the difference in functional attributes between the ith component and the jth component, so that the possibility of mutual substitution is smaller, and therefore the corresponding substitution coefficient is smaller; in addition, the greater the difference in size data between the ith component and the jth component, the less consistent the sizes of the ith component and the jth component, so that the possibility of mutual substitution is smaller, and therefore the corresponding substitution coefficient is smaller.

[0072] Further, the embodiment clusters each component in the professional cluster under each professional category with all the remaining components based on the substitution coefficient, and filters out the substitution cluster of each component under each professional category from all clusters based on the mean value of the substitution coefficient between all components in each cluster, specifically:

[0073] In the embodiment, each component in the professional cluster under each professional category and all the remaining components are taken as the input of the clustering algorithm, wherein the absolute value of the difference between the substitution coefficients is set as the metric distance of the clustering algorithm, the elbow method is used to determine the number of clustering clusters, and finally all clusters are output. In order to distinguish the clustering clusters in step S2, all the clusters in step S3 are collectively referred to as clusters, wherein the clustering algorithm adopts the k-means clustering algorithm.

[0074] Further, the average of the substitution coefficients between all components in each cluster is calculated, and the cluster with the maximum average of the substitution coefficients of all clusters of each component under each professional category is taken as the substitution cluster of each component under each professional category, which is used to represent a group of components in the same professional category that are very similar to a certain component in appearance size, use function and name. The components in the cluster have a high substitution coefficient between each other, which means that they are likely to be instances of the same basic component in different positions.

[0075] Preferably, the schematic diagram of the substitution cluster acquisition process provided by the embodiment is as shown in Figure 2

[0076] Further, the embodiment determines the retention coefficient of each component under each professional category by counting the number of directly connected components of each component and combining the difference in the substitution coefficients between each component under each professional category and all components in the substitution cluster of the component and the number of attributes, to obtain the remaining components of each component under each professional category, specifically as follows.

[0077] As a specific implementation, in the embodiment, the expression of the retention coefficient F u,v of component v under professional category u is as follows: L v represents the number of directly connected components of component v; N v represents the number of attributes of component v under professional category u; G u,v represents the difference in the substitution coefficients between component v under professional category u and all components in the substitution cluster of the component; and ε represents a preset constant greater than 0, which is used to prevent the denominator from being 0. In the embodiment, the value of ε is artificially set, and in the embodiment, the value of ε is 0.01. On the premise of ensuring that the denominator is not 0 and not excessively affecting the calculation result, the implementer can also set it according to the specific circumstances, and the embodiment does not make special limitations.

[0078] It should be noted that there are many methods for measuring the difference between data. In the embodiment, the absolute value of the difference in the substitution coefficients between component v under professional category u and all components in the substitution cluster of the component is taken as the difference in the substitution coefficients between component v under professional category u and all components in the substitution cluster of the component. In actual application, as other implementation manners, the implementer can also select other methods for measuring the difference between data, such as the square or ratio of the difference, and the embodiment does not make special limitations on the selection of the method for measuring the difference between data.

[0079] ​According to the retention coefficient of each component under each professional category, it can be understood that the retention coefficient is used to evaluate the importance of the component in the whole BIM model within the same professional category; the greater the number of components directly connected to component v, the more other components connected by component v in the structure of the BIM model, the more important the structural role of component v in the model, and the greater the impact on a larger range of other components when it is deleted, so it should be retained first, and the corresponding retention coefficient is relatively large; at the same time, the greater the number of attributes of component v under professional category u, the more attribute information contained by component v, which may carry more complex functions or have more detailed definitions, and the higher the information value, so it should be retained first, and the corresponding retention coefficient is relatively large; in addition, the smaller the difference between the substitution coefficients between component v and all components in its substitution cluster under professional category u, the smaller the average difference between component v and other components in the cluster, which means that component v is very representative in the cluster and has little essential difference from other components, and is the typical component in the cluster that should be retained first, therefore, the corresponding retention coefficient is larger;

[0080] On the contrary, the smaller the number of components directly connected to component v, the fewer other components connected by component v in the structure of the BIM model, the less important the structural role of component v in the model, and the smaller the impact on a smaller range of other components when it is deleted, so its priority for retention is relatively low, and the corresponding retention coefficient is relatively small; at the same time, the smaller the number of attributes of component v under professional category u, the less attribute information contained by component v, which may carry relatively simple functions or insufficient definitions, and the lower the information value, so its priority for retention is relatively low, and the corresponding retention coefficient is relatively small; in addition, the greater the difference between the substitution coefficients between component v and all components in its substitution cluster under professional category u, the greater the average difference between component v and other components in the cluster, which means that component v is more unique in the cluster and has certain essential difference from other components, and may not be the most representative typical component in the cluster, therefore, the corresponding retention coefficient is smaller.

[0081] Further, the component corresponding to the maximum retention coefficient in the substitution cluster of each component under each professional category is used as the retained component of each component under each professional category, which is used to replace the component that is very similar to the retained component in size, function and appearance, only with different spatial positions, so as to reduce the total number of components in the BIM model, realize the lightweight of the BIM model, and improve the efficiency of subsequent processing and checking.

[0082] So far, in this embodiment, the alternative clusters that can be replaced in the same specialty are identified through multi-specialty differentiation and similarity analysis, and the key remaining components are selected based on the structural importance, information value and representative of the components in the cluster, and the other components in the replacement cluster are represented by the remaining components, which not only ensures the integrity of the key information of the model, but also greatly reduces the model data, thereby significantly improving the checking efficiency of the multi-specialty BIM model of the complex road section.

[0083] Step S4: Lightening the BIM model based on the attribute data of all components under each remaining component in each specialty category to check the BIM model after lightening processing.

[0084] Based on step S3, the corresponding remaining component of each component is obtained, which means that a remaining component actually has one or more components, and the other similar components are represented by the remaining component. Therefore, in this embodiment, the BIM model is lightened based on the attribute data of all components under each remaining component in each specialty category to check the BIM model after lightening processing, thereby improving the efficiency of checking the multi-specialty design of the complex road section, specifically:

[0085] In this embodiment, all components under each remaining component in each specialty category are referred to as removed components, the attribute data of each removed component is referred to as an attribute set, the attribute set of the removed component is mapped to the remaining component, and all removed components of each remaining component in each specialty category are deleted, and all specialty categories are traversed to obtain the BIM model after lightening processing;

[0086] Further, the BIM model after lightening processing is taken as the input of the checking software, and the checking result is output.

[0087] It should be noted that if the number of attribute types of the remaining component is more than that of the removed component during the mapping process, the extra attributes of the remaining component are deleted and do not participate in the mapping.

[0088] It should be noted that there are many commonly used checking software, and the Dynamo software is used to check the BIM model after lightening processing in this embodiment. In actual application, as other implementation manners, the implementer can also use Navisworks, Revit and other software according to the specific situation, and the selection of the checking software is not specially limited in this embodiment.

[0089] So far, in this embodiment, the alternative clusters that can be replaced in the same specialty are identified through multi-specialty differentiation and similarity analysis, and the key remaining components are selected based on the structural importance, information value and representative of the components in the cluster, and the other components in the replacement cluster are represented by the remaining components, which not only ensures the integrity of the key information of the model, but also greatly reduces the model data, thereby significantly improving the checking efficiency of the multi-specialty BIM model of the complex road section.

[0090] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. And the above describes specific embodiments of the present specification. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0091] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between each of the embodiments can be referred to each other. Each of the embodiments focuses on the difference from other embodiments.

[0092] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit them; modifying the technical solutions recorded in the above-described embodiments, or equivalently replacing some of the technical features, does 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 BIM-based method for verifying multi-disciplinary design of complex road sections, characterized in that, The method includes the following steps: Obtain the names of all professional categories and the attribute data of each component in the BIM model of the viaduct project, and perform word segmentation on the names of each professional category and the attribute data of each component. In WordNet, traverse the root nodes of the hypernyms of each word segment; calculate the distances from the root nodes of all words in each component to the root nodes of all words in each professional category name to determine the first distance between each component and each professional category; obtain all synonyms of each word segment, analyze the differences of all synonyms between each component and each professional category name to determine the second distance between each component and each professional category, and form a distance vector with the first distance; cluster all components based on the distance vector, and filter out professional clusters under each professional category based on the magnitude of the distance vector corresponding to the cluster center in each cluster; The number of attributes and dimensional data for each component are statistically analyzed. The differences in the number of attributes and dimensional data between any two components are analyzed separately. Combined with the similarity of attribute data between any two components, the substitution coefficient between any two components is determined. Based on the substitution coefficient, each component in the professional cluster under each professional category is clustered with all other components. Based on the mean of the substitution coefficient between all components in each cluster, the substitution clusters for each component under each professional category are selected from all clusters. The number of components directly connected to each component is statistically analyzed. Combined with the differences in the substitution coefficient between each component under each professional category and all components in its substitution cluster, as well as the number of attributes, the retention coefficient for each component under each professional category is determined to obtain the retained components for each component under each professional category. The lightweight BIM model is based on the attribute data of all components under each retained component in each professional category, so as to verify the lightweight BIM model.

2. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, The first distance between each component and each professional category is the minimum value among the distances from the root node of all words of each component to the root node of all words of each professional category name.

3. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, The second distance between each component and each professional category is the word shift distance of all synonyms of all segments between each component and the name of each professional category.

4. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, In the process of clustering all components based on distance vectors, the distance is measured by the Euclidean distance between the distance vectors.

5. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, The professional clusters under each professional category are the clusters with the smallest distance vector magnitude corresponding to the cluster centers among all clusters under each professional category.

6. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, The expression for the substitution coefficient between any two components is: In the formula, A ij D represents the substitution coefficient between the i-th component and the j-th component; ij B represents the similarity of attribute data between the i-th component and the j-th component; ij C ij These represent the differences in the number of attributes and the differences in size data between the i-th component and the j-th component, respectively; τ represents a preset constant greater than 0.

7. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, The substitution cluster for each component under each professional category is the cluster with the largest mean substitution coefficient among all clusters of each component under each professional category.

8. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, The expression for the retention coefficient of each component under each professional category is as follows: F u,v L represents the retention factor of component v under professional category u; v N represents the number of components directly connected to component v; v G represents the number of attributes of component v under professional category u; u,v ε represents the difference in substitution coefficients between component v under professional category u and all components in its substitution cluster; ε represents a preset constant greater than 0.

9. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, The retained component for each component under each professional category is the component corresponding to the largest retention coefficient in the alternative cluster of each component under each professional category.

10. The BIM-based multi-disciplinary design verification method for complex road sections as described in claim 1, characterized in that, The lightweight BIM model, based on the attribute data of all components under each retained component in each professional category, is used to verify the lightweight BIM model, including: All components under each retained component in each professional category are recorded as removed components. The attribute data of each removed component are used to form an attribute set. The attribute set of the removed components is mapped to the retained components. All removed components of each retained component in each professional category are deleted. By traversing all professional categories, the lightweight BIM model is obtained. The lightweight BIM model is used as input to the verification software, and the verification results are output.

Images