CN110968713B - BIM multi-scale information retrieval method and device - Google Patents
BIM multi-scale information retrieval method and device Download PDFInfo
- Publication number
- CN110968713B CN110968713B CN201911168450.0A CN201911168450A CN110968713B CN 110968713 B CN110968713 B CN 110968713B CN 201911168450 A CN201911168450 A CN 201911168450A CN 110968713 B CN110968713 B CN 110968713B
- Authority
- CN
- China
- Prior art keywords
- building
- bim
- nodes
- matrix
- retrieval
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 239000011159 matrix material Substances 0.000 claims abstract description 172
- 230000011218 segmentation Effects 0.000 claims abstract description 35
- 238000010276 construction Methods 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 9
- 201000007946 Congenital intrinsic factor deficiency Diseases 0.000 claims description 4
- 208000034419 hereditary intrinsic factor deficiency Diseases 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 9
- 238000003058 natural language processing Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000009435 building construction Methods 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 239000002023 wood Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/51—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The embodiment of the invention provides a BIM multi-scale information retrieval method and device, wherein the method comprises the following steps: the method comprises the steps of segmenting query information of a building input by a user, obtaining a plurality of words, and performing semantic disambiguation on segmentation results; acquiring a feature matrix of the building and a search matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data; and obtaining a retrieval result corresponding to the query information according to the retrieval matrix and the feature matrix. The embodiment of the invention not only can search the information of the component level, but also can search the information of the attribute level, can better meet the requirements of users, and has simple search and high efficiency.
Description
Technical Field
The invention belongs to the technical field of building information retrieval, and particularly relates to a BIM multi-scale information retrieval method and device.
Background
The building information model (Building Information Modeling, BIM) integrates building information, such as space geometry information, building quantity and attribute information, related drawings, purchasing details, submitting flows and other construction files, and can be used for planning, designing, constructing and operating facilities, so that the traditional workflow and project delivery flows are changed, project cost is reduced, and production quality and efficiency are improved.
Multiscale information, namely component level information and attribute level information, is required at various stages of the building lifecycle. In one item, a user needs to retrieve not only a component in a building but also attribute information of the building. In the consistency check, the manager checks not only whether the components are satisfactory, but also whether the attribute values of the building, such as the width of the site, are satisfactory. In the construction phase, the user needs to extract construction-oriented information from the BIM design model to provide applications in the construction field. For example, structure-oriented takeoff information is extracted from the BIM design model. During the operation and maintenance phase, the equipment manager needs to extract information about various aspects of the building operation performance from the BIM model in order to operate and maintain the building. For example, information on aspects of construction operation performance is acquired. Research has shown that engineers are required to spend a great deal of time searching for information, and existing BIM retrieval systems have failed to meet the needs of users. It is therefore necessary to build a retrieval system capable of retrieving multi-scale information.
Retrieval techniques such as Autodesk Seek and BIMobject are widely used in the construction field. Research shows that the current construction information retrieval technology is inefficient in the syntactic analysis of complex sentences. Each building model contains a large number of concepts and terms, which place a cognitive burden on the user. Research has shown that even experts in the BIM field cannot retrieve the required information through several steps if the BIM model hierarchy is complex.
Disclosure of Invention
In order to overcome the problems of complex and low retrieval efficiency of the existing BIM multi-scale information retrieval method or at least partially solve the problems, the embodiment of the invention provides a BIM multi-scale information retrieval method and device.
According to a first aspect of an embodiment of the present invention, there is provided a BIM multi-scale information retrieval method, including:
the method comprises the steps of segmenting query information of a building input by a user, obtaining a plurality of words, and performing semantic disambiguation on segmentation results;
acquiring a feature matrix of the building and a search matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data;
and obtaining a retrieval result corresponding to the query information according to the retrieval matrix and the feature matrix.
Specifically, the step of obtaining the feature matrix of the building and the search matrix of the plurality of words according to the pre-constructed BIM hierarchical tree model further comprises the following steps:
acquiring a spatial structure of each building element in the BIM data according to an IfcRelContainedInSpatialStructure in an IFC standard of the BIM data;
Dividing the building elements into a plurality of layers according to the level of the space structure to which each building element belongs;
constructing the BIM hierarchical tree model according to the hierarchy to which the building element belongs; the nodes in the BIM hierarchical tree model comprise building nodes and attribute nodes, the building nodes represent building elements contained in parent nodes of the building nodes, and the attribute nodes represent attribute information of the parent nodes of the attribute nodes.
Specifically, the step of obtaining the feature matrix of the building according to the pre-constructed BIM hierarchical tree model comprises the following steps:
acquiring a retrieval path from a root node of the BIM hierarchical tree model to each node of a last hierarchy in the BIM hierarchical tree model; wherein the root node represents the building;
and determining the row of the feature matrix according to the building node and the attribute node of the building node on each retrieval path, and constructing the feature matrix by taking all the building nodes and the attribute nodes in the BIM hierarchical tree model as columns of the feature matrix.
Specifically, according to the building nodes and the attribute nodes of the building nodes on each path, determining the rows of the feature matrix, taking all the building nodes and the attribute nodes in the BIM hierarchical tree model as the columns of the feature matrix, and constructing the feature matrix specifically:
All building nodes and attribute nodes in the BIM hierarchical tree model are listed as follows:
A o ＝{(K 11 ,m 1 ,m 2 ,…,m n ),(K 21 ,m 1 ,m 2 ,…,m n ),…,(K in ,m 1 ,m 2 ,…,m n )}；
wherein A is o Represent the listed results, K in Representing the nth building node of the ith level in the BIM hierarchical tree model, wherein i is the total number of levels, K in N in (c) is the total number of building nodes of the ith hierarchy, (K) in ,m 1 ,m 2 ,…,m n ) M in (b) 1 ,m 2 ,…,m n Represent K in N attribute nodes of (a);
for any of the search paths, A o Building node allocation 1 on the search path, attribute node allocation 1 of the building node on the search path, and other building nodes and other attribute nodes allocation 0, and obtaining the row of the feature matrix corresponding to the search path.
Specifically, the step of obtaining the feature matrix of the building and the search matrix of the plurality of words according to the pre-constructed BIM hierarchical tree model further comprises the following steps:
acquiring GUIDs corresponding to the words and GUIDs corresponding to the nodes in the BIM hierarchical tree model from the IFDs;
judging whether node information which is the same as the GUID corresponding to each word exists or not according to the GUID corresponding to each word and the GUID corresponding to the information of each node;
if the word exists, replacing the word with node information which is the same as the GUID corresponding to the word;
If not, replacing the word with 0;
correspondingly, the step of obtaining the search matrix of the plurality of words according to the pre-constructed BIM hierarchical tree model comprises the following steps:
and obtaining the search matrix of the plurality of words after replacement according to a pre-constructed BIM hierarchical tree model.
Specifically, the step of obtaining the search matrix of the plurality of words after replacement according to the pre-constructed BIM hierarchical tree model includes:
is A o The node represented by the word after the replacement is allocated with 1;
is A o And allocating 0 to other nodes except the replaced nodes represented by the words, and obtaining the retrieval matrix.
Specifically, the step of obtaining the search result corresponding to the query information according to the search matrix and the feature matrix includes:
multiplying the feature matrix by the search matrix to obtain the number of all the searched nodes in each search path;
and determining the search result according to the search path corresponding to the maximum number of the nodes.
According to a second aspect of the embodiment of the present invention, there is provided a BIM multi-scale information retrieval apparatus, including:
the word segmentation module is used for segmenting query information of a building input by a user, acquiring a plurality of words and carrying out semantic disambiguation on segmentation results;
The acquisition module is used for acquiring the feature matrix of the building and the retrieval matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data;
and the retrieval module is used for acquiring a retrieval result corresponding to the query information according to the retrieval matrix and the feature matrix.
According to a third aspect of embodiments of the present invention, there is also provided an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor invoking the program instructions to be able to perform the BIM multi-scale information retrieval method provided by any of the various possible implementations of the first aspect.
According to a fourth aspect of embodiments of the present invention, there is also provided a non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the BIM multi-scale information retrieval method provided by any of the various possible implementations of the first aspect.
The embodiment of the invention provides a BIM multi-scale information retrieval method and a device, wherein the method is characterized in that by establishing a hierarchical relation of building information, BIH-Tree is used for representing BIM data in a layering way, a feature matrix of the BIH-Tree is constructed, semantic disambiguation is carried out after word segmentation is carried out on query information input by a user, a word-segmented retrieval matrix is constructed, and a retrieval result of the query information is obtained according to the retrieval matrix and the feature matrix, so that not only component-level information but also attribute-level information can be retrieved, and the requirements of the user can be better met; only the user needs to input query information once, so that the search is simple; BIH-Tree is used for representing BIM data in a layering manner, so that the retrieval efficiency is improved; the influence of natural language grammar and semantic complexity is reduced through semantic disambiguation, and a user can accurately obtain the required information only by describing a search target by using natural language.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic overall flow chart of a BIM multi-scale information retrieval method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a BIM hierarchical tree model in a BIM multi-scale information retrieval method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a search path in a BIM multi-scale information search method according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a BIH-Tree structure established in a BIM multi-scale information retrieval method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a result of component level information retrieval in a BIM multi-scale information retrieval method according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a result of attribute-level information retrieval in a BIM multi-scale information retrieval method according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of the overall structure of a BIM multi-scale information retrieval device according to an embodiment of the present invention;
fig. 8 is a schematic diagram of an overall structure of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
In one embodiment of the present invention, a BIM multi-scale information retrieval method is provided, and FIG. 1 is a schematic overall flow chart of the BIM multi-scale information retrieval method provided in the embodiment of the present invention, where the method includes S101, word segmentation is performed on query information of a building input by a user, a plurality of words are obtained, and semantic disambiguation is performed on word segmentation results;
specifically, when building information is retrieved in the BIM data, the user needs to input query information, where the query information is information related to a retrieval result and input by the user according to actual requirements. The query information is then segmented using a segmentation technique, which may be used with NLP (Natural Language Processing ) techniques, but is not limited to such segmentation techniques. Dividing query information input by a user into n>The 0 words, the word segmentation result includes one or more of nouns, adjectives, adverbs, numbers, prepositions, conjunctions and pronouns. All these words form a set, w= { W 1 ,w 2 ,…,w n Represented by w, where i (1.ltoreq.i.ltoreq.n) represents an i-th word obtained by word segmentation. The semantic disambiguation of the word segmentation result is to unify natural language input by the user with specialized terms in BIM data.
S102, acquiring a feature matrix of the building and a search matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data;
the present embodiment solves the problem of retrieval efficiency of the current BIM retrieval system by pre-establishing a hierarchical relationship between BIM data, for example, pre-establishing a hierarchical relationship between a second floor and a pillar before retrieving the pillar at the second floor of a building, i.e., the pillar belongs to the second floor. The searching range is limited to the second layer in advance before searching, so that the pillars can be directly searched in the second layer, and the information searching efficiency of BIM data is improved.
The present embodiment represents the hierarchical relationship between BIM data in a tree structure. A tree structure is a data structure that may be used to represent hierarchical relationships between building elements. BIM data can be partitioned into different hierarchies based on IFC (Industry Foundation Class, industry base class) standards and represented in a tree structure. The Tree structure model based on the IFC standard is defined as a BIM hierarchical Tree, which can be expressed as a BIH-Tree.
The Tree structure of the BIH-Tree is converted into a matrix form, and the converted matrix is used as a characteristic matrix of a building. Meanwhile, the word set is converted into a matrix according to the Tree structure of the BIH-Tree, and the matrix converted by the word set is used as a retrieval matrix. The present embodiment is not limited to the method of converting the tree structure and the word set into a matrix.
S103, obtaining a search result corresponding to the query information according to the search matrix and the feature matrix.
And obtaining a search result corresponding to the query information from the Tree structure of the BIH-Tree according to the search matrix and the feature matrix. The present embodiment is not limited to the manner of obtaining the search result.
In the embodiment, the hierarchical relation of building information is established, BIH-Tree is used for representing BIM data in a layering manner, a feature matrix of the BIH-Tree is established, after word segmentation is carried out on query information input by a user, a word segmentation retrieval matrix is established, and a retrieval result of the query information is obtained according to the retrieval matrix and the feature matrix, so that not only component-level information but also attribute-level information can be retrieved, and the requirements of the user can be better met; only the user needs to input query information once, so that the search is simple; BIH-Tree is used for representing BIM data in a layering manner, so that the retrieval efficiency is improved; the influence of natural language grammar and semantic complexity is reduced through semantic disambiguation, and a user can accurately obtain the required information only by describing a search target by using natural language.
Based on the above embodiment, the step of obtaining the feature matrix of the building and the search matrix of the plurality of words according to the pre-constructed BIM hierarchical tree model in this embodiment further includes: acquiring a spatial structure of each building element in the BIM data according to an IfcRelContainedInSpatialStructure in an IFC standard of the BIM data;
specifically, IFC standards incorporate ifcrlcontainedinsplatformstruct to assign architectural elements to specific levels of a spatial structure. Any building element can only be allocated once at a specific level of the spatial structure. Furthermore, building elements must be layered and contained only in a single spatial structure. One architectural element can reference many spatial structures, and the reference relationship is represented by ifcrelcontainedinsplatatialstructure. Predefined spatial structures, such as IfcSite, ifcBuilding, ifcBuildingStorey and IfcSpace, may be allocated for building elements. The same building element type may be assigned to different spatial structures depending on the context of the event.
Dividing the building elements into a plurality of layers according to the level of the space structure to which each building element belongs;
BIH-Tree is a hierarchical representation Tree of building information, with the root node of BIH-Tree representing the overall model of the building. Under the root node, BIH-Tree may be classified as i > 0. Each level of the BIH-Tree includes n >0 nodes.
Constructing the BIM hierarchical tree model according to the hierarchy to which the building element belongs; the nodes in the BIM hierarchical tree model comprise building nodes and attribute nodes, the building nodes represent building elements contained in parent nodes of the building nodes, and the attribute nodes represent attribute information of the parent nodes of the attribute nodes.
As shown in fig. 2, the nodes to which the solid lines are connected in the BIM hierarchical tree model are defined as building nodes, called BN. BN is denoted as K o ＝{K i1 ,K i2 ,…,K in }。K ij The method is characterized in that the method comprises the steps of representing the j-th building node of the i-th level in the BIH-Tree, wherein i is more than or equal to 1 and less than or equal to n, j is more than or equal to 1 and less than or equal to m, n is the total level number, and m is the number of nodes on the i-th level. Since each building node in the BIH-Tree contains n>0 pieces of attribute information connected by a dotted line and m= { M 1 ,m 2 ,…,m n Represented by (m) i And the i-th attribute information of the building node is represented, wherein i is more than or equal to 1 and less than or equal to n. The node to which the dotted line is connected is defined as AN attribute node, called AN. Two nodes, such as BN and BN, or BN and AN, connected by solid or dashed lines in fig. 2 represent hierarchical relationships of information in BIM. For example, K 21 Represents K 11 Sub-building elements, K 11 M to 1 Representative ofBuilding node K 11 I.e., the attribute node of the i+1 layer represents the attribute node of the parent construction node of the i layer. Thereby representing the hierarchical relationship of the BIM data as a BIM hierarchical tree.
The whole construction process of the BIH-Tree comprises the following steps:
1. determining building nodes and attribute nodes according to building elements and attribute information in BIM data;
2. determining a root node of the BIH-Tree, and taking the root node as a first level of the BIH-Tree;
3. acquiring a sub-building node belonging to a root node and an attribute node of the root node, connecting the root node and the sub-building node belonging to the root node by a solid line, connecting the attribute node of the root node and the attribute node of the root node by a broken line, and constructing a second level of the BIH-Tree;
4. and continuing to acquire the sub-building nodes belonging to the building nodes in the second level and the attribute nodes of the building nodes in the second level, and connecting until the sub-building nodes of all the building nodes in the current level can not be found.
On the basis of the above embodiment, the step of obtaining the feature matrix of the building according to the pre-constructed BIM hierarchical tree model in this embodiment includes: acquiring a retrieval path from a root node of the BIM hierarchical tree model to each node of a last hierarchy in the BIM hierarchical tree model; wherein the root node represents the building; and determining the row of the feature matrix according to the building node and the attribute node of the building node on each retrieval path, and constructing the feature matrix by taking all the building nodes and the attribute nodes in the BIM hierarchical tree model as columns of the feature matrix.
Specifically, assume that the number of building nodes in the BIH-Tree is m, and each building node has n attribute nodes. The combination of the building node and the attribute node on the path from the root node to the attribute node of the last hierarchy is defined as a retrieval path P, as shown in fig. 3. Assuming that the number of nodes on the last hierarchy is N, the total number of retrieval paths is N. Search route p= { P 1 ,p 2 ,…,p N Represented by p i The path is retrieved for the ith. And then listing the nodes on the BIH-Tree according to the sequence of the building nodes and the attribute nodes of the building nodes through which the path passes. For example, the first path in FIG. 3 may be written as p 1 ＝{K 11 ,m 1 ,m 2 ,…,m n ,K 21 ,m 1 ,m 2 ,…,m n ,K 31 ,m 1 Comprises 4 search paths, i.e. p, as in FIG. 3 1 、p 2 、p 3 And p 4 。
Therefore, all nodes in BIH-Tree are listed in terms of building nodes and their attribute node combinations, namely A o ＝{(K 11 ,m 1 ,m 2 ,…,m n ),(K 21 ,m 1 ,m 2 ,…,m n ),…,(K in ,m 1 ,m 2 ,…,m n )}。A o The total number of nodes in (a) is n+nxn, i.e., n× (n+1). According to A o A matrix representation of each retrieval path is determined. A is that o Represent the listed results, K in Representing the nth building node of the ith level in the BIM hierarchical tree model, i being the total number of levels, m being the total number of building nodes of the ith level, (K) in ,m 1 ,m 2 ,…,m n ) M in (b) 1 ,m 2 ,…,m n Represent K in N attribute nodes of (a);
then, the N search paths are taken as rows of the matrix, and all building nodes and attribute nodes contained in the BIH-Tree are taken as columns of the matrix. For any of the search paths p i Is A o Building node K on the search path ij Distribution 1, A o Attribute node m of the building node existing on the search path i Dispensing 1, A o And (3) distributing 0 to other building nodes and other attribute nodes in the search path to acquire the rows of the feature matrix corresponding to the search path, namely:
according to formula (1), a feature matrix A of BIH-Tree can be obtained. The BIH-Tree feature matrix A for N rows and n× (n+1) columns is shown below:
based on the above embodiment, the step of obtaining the feature matrix of the building and the search matrix of the plurality of words according to the pre-constructed BIM hierarchical tree model in this embodiment further includes: acquiring GUIDs corresponding to the words and GUIDs corresponding to the nodes in the BIM hierarchical tree model from the IFDs; judging whether node information which is the same as the GUID corresponding to each word exists or not according to the GUID corresponding to each word and the GUID corresponding to the information of each node; if the word exists, replacing the word with node information which is the same as the GUID corresponding to the word; if not, replacing the word with 0; correspondingly, the step of obtaining the search matrix of the plurality of words according to the pre-constructed BIM hierarchical tree model comprises the following steps: and obtaining the search matrix of the plurality of words after replacement according to a pre-constructed BIM hierarchical tree model.
Specifically, when building information is retrieved in the BIM data, the user must accurately input query information. Otherwise, it is difficult to retrieve the desired information. This brings great inconvenience to the user. With the continued advancement of computer technology, there is an increasing search for communication with computers using natural language. To achieve this goal, the present embodiment requires that the query be parsed using NLP to separate the query information into multiple words.
Because of the complexity and ambiguity of natural language, there are many ways in which the same building information, i.e., one meaning multiple words, can be described. For example, "student dormitory" and "student apartment" are used to describe a building in which students live. When the user describes the same objective thing, a rich description form will be generated due to the differences of country, region, language, culture, etc. Obviously, due to the above factors, the existing information retrieval system cannot analyze different expressions of the same building information, and thus cannot retrieve corresponding building information.
The present embodiment uses IFD (International Framework for Dictionary, international dictionary framework) to unify words, i.e., to perform semantic disambiguation. IFDs semantically describe some concepts and terms and define each architectural concept as a unique GUID in the world to ensure uniformity of information. To provide accurate building information for information exchange, the IFD tags all information in the IFC format with GUID, providing a mapping method from concepts to IFC entities. Therefore, different languages, synonyms and the like can be mapped to the same IFC entity, and a foundation is laid for establishing information interaction. Building information provider only provides corresponding GUID of building in the process of information exchange, thereby eliminating ambiguity caused by language and culture.
IFD is used to disambiguate each word obtained. Defining an IFD-based building information dictionary comprising n>0 words, and is represented by d= { D 1 ,d 2 ,…,d n Represented by d, where i Representing the ith building information in dictionary D, their unique corresponding GUIDs can be found in dictionary D, thereby disambiguating. If some words in W cannot represent nodes on BIH-Tree, disambiguation is not required. Finally, n is obtained>0 words, which have been semantically disambiguated. These semantically disambiguated words are represented by k= { K 1 ,k 2 ,…,k n Represented by k j (1. Ltoreq.j.ltoreq.n) is the j-th word after semantic elimination. Each step based on IFD semantic disambiguation can be written as:
the above expression indicates that if the i-th word w after word segmentation i Corresponds to the jth word k in the IFD-based dictionary D j Will w i Replaced by k j Otherwise w i The value of (2) returns to 0.
According to the method and the device, the query is analyzed and unified by utilizing the natural language processing technology NLP and the international dictionary framework IFD, namely word segmentation and semantic disambiguation, the influence of natural language grammar and semantic complexity is reduced, and a user can accurately obtain the required information by describing a search target by using natural language.
Based on the above embodiment, the step of obtaining the search matrix of the plurality of words after replacement according to the pre-constructed BIM hierarchical tree model in this embodiment includes: is A o The node represented by the word after the replacement is allocated with 1; is A o And allocating 0 to other nodes except the replaced nodes represented by the words, and obtaining the retrieval matrix.
Specifically, the set K obtained through query analysis and unification is in the set A of BIH-Tree o Is traversed in the middle. In general, when a user searches for a building element having a specific attribute or acquires attribute information of a certain building element, it is the attribute information of a building node that is closest to the building node, and thus the closest traversal rule is followed. Each key in K follows the traversal order from the building node to the attribute nodes connected thereto and traverses the attribute node in the nearest building node. Is A o The node represented by the word after the replacement is allocated with 1; is A o And allocating 0 to other nodes except the replaced node represented by the word to acquire a retrieval matrix. The search matrix B is obtained from the following relation:
according to formula (4), the search matrix B can be obtained. The row representation with element 1 of search matrix B may be in set A o The parsed and unified word found in the database. The search matrix B of n× (n+1) rows and one column is shown below.
Based on the foregoing embodiments, in this embodiment, the step of obtaining, according to the search matrix and the feature matrix, a search result corresponding to the query information includes: multiplying the feature matrix by the search matrix to obtain the number of nodes searched in each search path; and determining the search result according to the search path corresponding to the maximum number of the nodes.
Specifically, the obtained BIH-Tree feature matrix A is multiplied by the search matrix B to obtain a matrix C, as follows:
the feature matrix a and the search matrix B are multiplied to obtain a matrix of N rows and one column, which is represented by c= [ p ] 1 o ,p 2 o ,...,p N o ] T And (3) representing. Each row of matrix C represents the algebraic sum of all nodes that can be retrieved in each retrieval path. Obviously, the maximum number in C is the query result, so the elements in matrix C are ordered from high to low to obtain a new matrix, by C o And (3) representing. C (C) o The first element in (a) is the search result.
In the embodiment, node information in different retrieval paths is traversed by constructing a BIH-Tree feature matrix A, a retrieval matrix B and a retrieval path P to obtain a matrix C, and then the matrix C is ordered to obtain a retrieval result matrix C o . The search result is C o Node information on a retrieval path corresponding to the most numerical element, thereby realizing the retrieval of multi-scale information. Can help the related personnel of the project to acquire the required information more conveniently and accurately. The method may also be used for facility management, cost management, and other applications.
Taking a library of a university as an example, the feasibility of the BIM multi-scale information retrieval method using hierarchical modeling and natural language processing in the embodiment is verified. The library is a square building with the height of 32m, the length of 70m and the width of 70m, and the total building area is 38580m 2 3.2 billions of yuan are consumed. It comprises seven layers on the ground and one layer underground. Each layer is divided into four regions, a, B, C and D.
According to the IFC standard, a hierarchical relationship between building information is established, and a building information model is established. The method comprises the following specific steps:
1. BIM models of the library were constructed using Autodesk Revit 2016. The model includes building information in the areas of building construction, water supply and drainage, heating and ventilation, and electricity.
2. Each floor of the building is divided into four zones, namely, zone a, zone B, zone C and zone D, according to the structural characteristics of the building and IFC criteria. Some buildings are allocated in these areas.
3. And building the BIH tree according to the hierarchical relationship and the IFC standard among different buildings in the same system. BIH-Tree is imported into each building unit and a hierarchical building information model is designed. For example, the bookshelf belongs to the B area of the second floor of the library, and then the BIH-Tree containing the bookshelf is established and imported into the attribute information of the bookshelf to complete the design of the model.
4. And importing all BIH-Tree data representing the hierarchical relationship into a retrieval system to realize the function of searching building components associated with the unified query.
The entire library building consists of seven floors. Each floor is divided into four zones, stored in IfcSpace, e.g., zone a and zone B, respectively. All building elements in a building are assigned to different levels within the spatial structure according to the ifcrelcontainedinsplatatual structure relationship. For example: stairs are assigned to buildings, walls are assigned to floors, and shelves are assigned to spaces. All attribute information of the building elements is integrated as data into each building element. In this experiment, 10 out of n building elements were selected as building nodes, respectively "library", "two-layer", "three-layer", "region a", "region B", "region C", "fire detector", "hydrant", "bookshelf" and "desk". 8 attribute nodes are selected from n×n attribute information, and are respectively ' 3.2 hundred million yuan ', ' total area: 5511m2"," area: 1377m2"," color: white "," material: plastic "," area: 0.75m2"," 200-membered "," material: wood. In the embodiment, the 18 nodes are taken as experimental samples, and component-level information retrieval and attribute-level information retrieval are realized in the BIM.
Component level information retrieval is retrieving components in the BIM model. The retrieved results will return a component and visualize the results. According to the method of building BIH-Tree, 18 experimental samples were used to build BIH-Tree, as shown in FIG. 4. The construction of the BIH-Tree feature matrix A is shown as follows:
then, the experiment uses the "jieba" word segmentation module in Python for word segmentation. The module can customize the dictionary according to the own needs so as to improve the accuracy of word segmentation. The experiment uses the "jieba. Cut_for_search" function to word the query. The query information is represented by w= { a white fire alarm located in the two-floor a area of the library }. After word segmentation, a group of words is obtained, namely W= { W 1 = "located", w 2 = "library", w 3 = "two layers", w 4 = "a region", w 5 = "middle", w 6 = "w 7 = "one", w 8 = "white", w 9 = "w 10 = "fire alarm" }.
Then, performing semantic disambiguation on the elements in W according to the dictionary D to obtain a group of words with unified standards, namely K= { K 1 = "at", k 2 = "library", k 3 = "two layers", k 4 = "region a", k 5 = "middle", k 6 = "k 7 = "one", k 8 = "white", k 9 = "k 10 = "fire detector" }.
Then, K is taken up 1 Written in the form of a search matrix B, namely:
multiplying the BIH-Tree feature matrix A by the search matrix B to obtain a component search result matrix C, namely:
the elements in matrix C are then ordered from high to low to obtain a final search result matrix C o ＝[5 4 3 2 1] T . Obviously C o Is the building information required by the user. Thus, the search results are all nodes searched in the first path, as indicated by the dotted-line circled portion in fig. 5.
In actual retrieval, the user may not enter a particular building node. In this case, the search result may be obtained by searching for child nodes or attribute nodes of the construction node. In other words, the search method of the present embodiment can search for multi-scale building information, i.e., any level of building information in the BIH-Tree.
The user may retrieve building elements in the model of the library. First, a natural language describing a search target is input in a search box, then a query button is clicked to acquire a search result, and the searched member is displayed in three dimensions. When clicking on the retrieved building element, the attribute information is reviewed and the element name, element type, parent element, GUID and other element information of the retrieved building element is displayed.
Attribute level information retrieval retrieves attribute information of a building according to item requirements, and a retrieval result returns an attribute value instead of a component. Query information w= { total area of library second floor }. After word segmentation, a group of words is obtained, namely W= { W 1 = "library", w 2 = "second floor", w 3 = "w 4 = "total area" }.
Then, performing semantic disambiguation on the elements in W according to the dictionary D to obtain a group of words with unified standards, namely K= { K 1 = "library", k 2 = "two layers", k 3 = "k 4 = "total area" }. Then, according to the above closest traversal rule, K can be expressed as the following search matrix B form.
The attribute result matrix C is obtained by multiplying the BIH-Tree feature matrix a and the attribute retrieval matrix B as follows:
then the element matrix C is ordered from high to low to obtain a search result matrix C o ＝[3 3 3 3 1] T . Obviously C o The first location element in (a) is search information required by the user, and the searched node visually displays the searched floor attribute information as shown by a part circled by a dotted line in fig. 6.
The results of component level and attribute level retrieval indicate that the multi-scale building information retrieval method proposed by the embodiment is feasible. The multi-scale building information retrieval method allows non-professional users to retrieve building information at a component level or attribute level using natural language, thereby improving the data value of BIM.
In another embodiment of the present invention, a BIM multi-scale information retrieval apparatus is provided for implementing the method of the previous embodiments. Therefore, the descriptions and definitions in the embodiments of the BIM multi-scale information retrieval method described above can be used for understanding the respective execution modules in the embodiments of the present invention. Fig. 7 is a schematic diagram of the overall structure of a BIM multi-scale information retrieval device according to an embodiment of the present invention, where the device includes a word segmentation module 701, an obtaining module 702, and a retrieval module 703; wherein:
the word segmentation module 701 is configured to segment query information of a building input by a user, obtain a plurality of words, and perform semantic disambiguation on a segmentation result;
when building information is searched in BIM data, a user needs to input query information, wherein the query information is information related to a search result and input by the user according to actual requirements. The word segmentation module 701 uses word segmentation technology to segment the query information, and NLP technology may be used to segment the query information, but is not limited to this word segmentation technology. The semantic disambiguation of the word segmentation result is to unify natural language input by the user with specialized terms in BIM data.
The obtaining module 702 is configured to obtain a feature matrix of the building and a search matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data;
The hierarchical relationship between BIM data is represented by a tree structure. A tree structure is a data structure that may be used to represent hierarchical relationships between building elements. BIM data may be divided into different hierarchies based on IFC standards and represented in a tree structure. The Tree structure model based on the IFC standard is defined as a BIM hierarchical Tree, which can be expressed as a BIH-Tree.
The obtaining module 702 converts the Tree structure of the BIH-Tree into a matrix form, and uses the converted matrix as a feature matrix of the building. Meanwhile, the word set is converted into a matrix according to the Tree structure of the BIH-Tree, and the matrix converted by the word set is used as a retrieval matrix. The present embodiment is not limited to the method of converting the tree structure and the word set into a matrix.
The retrieving module 703 is configured to obtain a retrieving result corresponding to the query information according to the retrieving matrix and the feature matrix.
The retrieval module 703 obtains a retrieval result corresponding to the query information from the Tree structure of the BIH-Tree according to the retrieval matrix and the feature matrix. The present embodiment is not limited to the manner of obtaining the search result.
In the embodiment, the hierarchical relation of building information is established, BIH-Tree is used for representing BIM data in a layering manner, a feature matrix of the BIH-Tree is established, after word segmentation is carried out on query information input by a user, a word segmentation retrieval matrix is established, and a retrieval result of the query information is obtained according to the retrieval matrix and the feature matrix, so that not only component-level information but also attribute-level information can be retrieved, and the requirements of the user can be better met; only the user needs to input query information once, so that the search is simple; BIH-Tree is used for representing BIM data in a layering manner, so that the retrieval efficiency is improved; the influence of natural language grammar and semantic complexity is reduced through semantic disambiguation, and a user can accurately obtain the required information only by describing a search target by using natural language.
On the basis of the above embodiment, the present embodiment further includes a construction module, configured to obtain, according to ifcrelcontainedinpatialstructure in the IFC standard of the BIM data, a spatial structure to which each building element in the BIM data belongs; dividing the building elements into a plurality of layers according to the level of the space structure to which each building element belongs; constructing the BIM hierarchical tree model according to the hierarchy to which the building element belongs; the nodes in the BIM hierarchical tree model comprise building nodes and attribute nodes, the building nodes represent building elements contained in parent nodes of the building nodes, and the attribute nodes represent attribute information of the parent nodes of the attribute nodes.
On the basis of the above embodiment, the acquiring module in this embodiment is specifically configured to: acquiring a retrieval path from a root node of the BIM hierarchical tree model to each node of a last hierarchy in the BIM hierarchical tree model; wherein the root node represents the building; and determining the row of the feature matrix according to the building node and the attribute node of the building node on each retrieval path, and constructing the feature matrix by taking all the building nodes and the attribute nodes in the BIM hierarchical tree model as columns of the feature matrix.
On the basis of the above embodiment, the obtaining module in this embodiment is further configured to list all building nodes and attribute nodes in the BIM hierarchical tree model in the following manner:
A o ＝{(K 11 ,m 1 ,m 2 ,…,m n ),(K 21 ,m 1 ,m 2 ,…,m n ),…,(K in ,m 1 ,m 2 ,…,m n )}；
wherein A is o Represent the listed results, K in Representing the nth building node of the ith level in the BIM hierarchical tree model, wherein i is the total number of levels, K in N in (c) is the total number of building nodes of the ith hierarchy, (K) in ,m 1 ,m 2 ,…,m n ) M in (b) 1 ,m 2 ,…,m n Represent K in N attribute nodes of (a);
for any of the search paths, A o Building node allocation 1 on the search path, attribute node allocation 1 of the building node on the search path, and other building nodes and other attribute nodes allocation 0, and obtaining the row of the feature matrix corresponding to the search path.
Based on the above embodiment, the present embodiment further includes a preprocessing module, configured to obtain, from an IFD, a GUID corresponding to each word and a GUID corresponding to each node in the BIM hierarchical tree model; judging whether node information which is the same as the GUID corresponding to each word exists or not according to the GUID corresponding to each word and the GUID corresponding to the information of each node; if the word exists, replacing the word with node information which is the same as the GUID corresponding to the word; if not, replacing the word with 0; correspondingly, the acquisition module is specifically configured to: and obtaining the search matrix of the plurality of words after replacement according to a pre-constructed BIM hierarchical tree model.
On the basis of the above embodiment, the acquiring module in this embodiment is specifically configured to: is A o The node represented by the word after the replacement is allocated with 1; is A o And allocating 0 to other nodes except the replaced nodes represented by the words, and obtaining the retrieval matrix.
On the basis of the above embodiments, the search module in this embodiment is specifically configured to: multiplying the feature matrix and the search matrix to obtain the number of all the searched nodes in each search path; and determining the search result according to the search path corresponding to the maximum number of the nodes.
Fig. 8 illustrates a physical structure diagram of an electronic device, as shown in fig. 8, which may include: a processor 801, a communication interface (Communications Interface) 802, a memory 803, and a communication bus 804, wherein the processor 801, the communication interface 802, and the memory 803 communicate with each other through the communication bus 804. The processor 801 may call logic instructions in the memory 803 to perform the following method: the method comprises the steps of segmenting query information of a building input by a user, obtaining a plurality of words, and performing semantic disambiguation on segmentation results; acquiring a feature matrix of the building and a search matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data; and obtaining a retrieval result corresponding to the query information according to the retrieval matrix and the feature matrix.
Further, the logic instructions in the memory 803 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The present embodiment provides a non-transitory computer readable storage medium storing computer instructions that cause a computer to perform the methods provided by the above-described method embodiments, for example, including: the method comprises the steps of segmenting query information of a building input by a user, obtaining a plurality of words, and performing semantic disambiguation on segmentation results; acquiring a feature matrix of the building and a search matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data; and obtaining a retrieval result corresponding to the query information according to the retrieval matrix and the feature matrix.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware associated with program instructions, where the foregoing program may be stored in a computer readable storage medium, and when executed, the program performs steps including the above method embodiments; and the aforementioned storage medium includes: various media that can store program code, such as ROM, RAM, magnetic or optical disks.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (7)
1. A BIM multi-scale information retrieval method, comprising:
the method comprises the steps of segmenting query information of a building input by a user, obtaining a plurality of words, and performing semantic disambiguation on segmentation results;
acquiring a feature matrix of the building and a search matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data;
obtaining a retrieval result corresponding to the query information according to the retrieval matrix and the feature matrix;
The step of obtaining the feature matrix of the building and the search matrix of the plurality of words according to the pre-constructed BIM hierarchical tree model further comprises the following steps:
acquiring a spatial structure of each building element in the BIM data according to an IfcRelContainedInSpatialStructure in an IFC standard of the BIM data;
dividing the building elements into a plurality of layers according to the level of the space structure to which each building element belongs;
constructing the BIM hierarchical tree model according to the hierarchy to which the building element belongs; the nodes in the BIM hierarchical tree model comprise building nodes and attribute nodes, wherein the building nodes represent building elements contained in father nodes of the building nodes, and the attribute nodes represent attribute information of the father nodes of the attribute nodes;
the step of obtaining the feature matrix of the building according to the pre-constructed BIM hierarchical tree model comprises the following steps:
acquiring a retrieval path from a root node of the BIM hierarchical tree model to each node of a last hierarchy in the BIM hierarchical tree model; wherein the root node represents the building;
determining the rows of the feature matrix according to the building nodes on each retrieval path and the attribute nodes of the building nodes, and constructing the feature matrix by taking all the building nodes and the attribute nodes in the BIM hierarchical tree model as columns of the feature matrix;
Determining the row of the feature matrix according to the building nodes on each path and the attribute nodes of the building nodes, taking all the building nodes and the attribute nodes in the BIM hierarchical tree model as columns of the feature matrix, and constructing the feature matrix specifically:
all building nodes and attribute nodes in the BIM hierarchical tree model are listed as follows:
A o ＝{(K 11 ,m 1 ,m 2 ,…,m n ),(K 21 ,m 1 ,m 2 ,…,m n ),…,(K in ,m 1 ,m 2 ,…,m n )}；
wherein A is o Represent the listed results, K in Representing the nth building node of the ith level in the BIM hierarchical tree model, wherein i is the total number of levels, K in N in (c) is the total number of building nodes of the ith hierarchy, (K) in ,m 1 ,m 2 ,…,m n ) M in (b) 1 ,m 2 ,…,m n Represent K in N attribute nodes of (a);
for any of the search paths, A o Building node allocation 1 on the search path, attribute node allocation 1 of the building node on the search path, and other building nodes and other attribute nodes allocation 0, and obtaining the row of the feature matrix corresponding to the search path.
2. The BIM multi-scale information retrieval method according to claim 1, wherein the step of obtaining the feature matrix of the building and the retrieval matrix of the plurality of words from the pre-constructed BIM hierarchical tree model is preceded by the steps of:
Acquiring GUIDs corresponding to the words and GUIDs corresponding to the nodes in the BIM hierarchical tree model from the IFDs;
judging whether node information which is the same as the GUID corresponding to each word exists or not according to the GUID corresponding to each word and the GUID corresponding to the information of each node;
if the word exists, replacing the word with node information which is the same as the GUID corresponding to the word;
if not, replacing the word with 0;
correspondingly, the step of obtaining the search matrix of the plurality of words according to the pre-constructed BIM hierarchical tree model comprises the following steps:
and obtaining the search matrix of the plurality of words after replacement according to a pre-constructed BIM hierarchical tree model.
3. The BIM multi-scale information retrieval method according to claim 2, wherein the step of obtaining the retrieval matrix of the plurality of words after the replacement according to the pre-constructed BIM hierarchical tree model includes:
is A o The node represented by the word after the replacement is allocated with 1;
is A o And allocating 0 to other nodes except the replaced nodes represented by the words, and obtaining the retrieval matrix.
4. A BIM multi-scale information retrieval method according to any one of claims 1 to 3, wherein the step of obtaining the retrieval result corresponding to the query information according to the retrieval matrix and the feature matrix includes:
Multiplying the feature matrix by the search matrix to obtain the number of all the searched nodes in each search path;
and determining the search result according to the search path corresponding to the maximum number of the nodes.
5. A BIM multi-scale information retrieval apparatus, comprising:
the word segmentation module is used for segmenting query information of a building input by a user, acquiring a plurality of words and carrying out semantic disambiguation on segmentation results;
the acquisition module is used for acquiring the feature matrix of the building and the retrieval matrix of the plurality of words according to a pre-constructed BIM hierarchical tree model; the BIM hierarchical tree model is a model for dividing BIM data of the building into a plurality of layers and representing the BIM data of the building into a tree structure according to the layers of the BIM data;
the retrieval module is used for acquiring a retrieval result corresponding to the query information according to the retrieval matrix and the feature matrix;
the device also comprises a construction module for:
acquiring a spatial structure of each building element in the BIM data according to an IfcRelContainedInSpatialStructure in an IFC standard of the BIM data;
dividing the building elements into a plurality of layers according to the level of the space structure to which each building element belongs;
Constructing the BIM hierarchical tree model according to the hierarchy to which the building element belongs; the nodes in the BIM hierarchical tree model comprise building nodes and attribute nodes, wherein the building nodes represent building elements contained in father nodes of the building nodes, and the attribute nodes represent attribute information of the father nodes of the attribute nodes;
the acquisition module is used for:
acquiring a retrieval path from a root node of the BIM hierarchical tree model to each node of a last hierarchy in the BIM hierarchical tree model; wherein the root node represents the building;
determining the rows of the feature matrix according to the building nodes on each retrieval path and the attribute nodes of the building nodes, and constructing the feature matrix by taking all the building nodes and the attribute nodes in the BIM hierarchical tree model as columns of the feature matrix;
the acquisition module is used for:
all building nodes and attribute nodes in the BIM hierarchical tree model are listed as follows:
A o ＝{(K 11 ,m 1 ,m 2 ,…,m n ),(K 21 ,m 1 ,m 2 ,…,m n ),…,(K in ,m 1 ,m 2 ,…,m n )}；
wherein A is o Represent the listed results, K in Representing the nth building node of the ith level in the BIM hierarchical tree model, wherein i is the total number of levels, K in N in (c) is the total number of building nodes of the ith hierarchy, (K) in ,m 1 ,m 2 ,…,m n ) M in (b) 1 ,m 2 ,…,m n Represent K in N attribute nodes of (a);
for any of the search paths, A o Building node allocation 1 on the search path, attribute node allocation 1 of the building node on the search path, and other building nodes and other attribute nodes allocation 0, and obtaining the row of the feature matrix corresponding to the search path.
6. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor performs the steps of the BIM multiscale information retrieval method according to any one of claims 1 to 4 when the program is executed.
7. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the BIM multiscale information retrieval method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911168450.0A CN110968713B (en) | 2019-11-25 | 2019-11-25 | BIM multi-scale information retrieval method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911168450.0A CN110968713B (en) | 2019-11-25 | 2019-11-25 | BIM multi-scale information retrieval method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110968713A CN110968713A (en) | 2020-04-07 |
| CN110968713B true CN110968713B (en) | 2023-05-12 |
Family
ID=70031563
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911168450.0A Active CN110968713B (en) | 2019-11-25 | 2019-11-25 | BIM multi-scale information retrieval method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110968713B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111428381A (en) * | 2020-04-14 | 2020-07-17 | 中国交通建设股份有限公司吉林省分公司 | Building BIM data and evaluation data conversion system, method, equipment and storage medium |
| CN112905591B (en) * | 2021-02-04 | 2022-08-26 | 成都信息工程大学 | Data table connection sequence selection method based on machine learning |
| CN116069834B (en) * | 2023-03-07 | 2023-06-23 | 武汉东方赛思软件股份有限公司 | BIM technology-based enterprise data visual management method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2629210A1 (en) * | 2012-02-20 | 2013-08-21 | Solideo Systems Co., Ltd. | Method of providing data included in building information modeling data file, computer readable medium therefor, system using the method, and method of providing data using building information modeling server |
| CN103500371A (en) * | 2013-08-08 | 2014-01-08 | 机械工业第六设计研究院有限公司 | Change control method based on BIM |
| CN107247757A (en) * | 2017-05-31 | 2017-10-13 | 上海建工四建集团有限公司 | Construction project document classification storage and retrieval method and system based on BIM |
| CN109190094A (en) * | 2018-09-05 | 2019-01-11 | 盈嘉互联（北京）科技有限公司 | Building Information Model file cutting method based on IFC standard |
| CN110188454A (en) * | 2019-05-27 | 2019-08-30 | 北京建筑大学 | Architectural Equipment and Building Information Model matching process and device |
-
2019
- 2019-11-25 CN CN201911168450.0A patent/CN110968713B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2629210A1 (en) * | 2012-02-20 | 2013-08-21 | Solideo Systems Co., Ltd. | Method of providing data included in building information modeling data file, computer readable medium therefor, system using the method, and method of providing data using building information modeling server |
| CN103500371A (en) * | 2013-08-08 | 2014-01-08 | 机械工业第六设计研究院有限公司 | Change control method based on BIM |
| CN107247757A (en) * | 2017-05-31 | 2017-10-13 | 上海建工四建集团有限公司 | Construction project document classification storage and retrieval method and system based on BIM |
| CN109190094A (en) * | 2018-09-05 | 2019-01-11 | 盈嘉互联（北京）科技有限公司 | Building Information Model file cutting method based on IFC standard |
| CN110188454A (en) * | 2019-05-27 | 2019-08-30 | 北京建筑大学 | Architectural Equipment and Building Information Model matching process and device |
Non-Patent Citations (1)
| Title |
|---|
| "基于知识重用的建筑铝模板智能设计技术研究与系统实现";苗志民;《中国优秀硕士学位论文全文数据库 工程科技》;20190715;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110968713A (en) | 2020-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108288229B (en) | User portrait construction method | |
| CN110968713B (en) | BIM multi-scale information retrieval method and device | |
| US20160103837A1 (en) | System for, and method of, ranking search results obtained by searching a body of data records | |
| Zhang et al. | Ontology-based semantic retrieval for engineering domain knowledge | |
| US20160103885A1 (en) | System for, and method of, building a taxonomy | |
| US9672279B1 (en) | Cluster labeling system for documents comprising unstructured text data | |
| US20160103920A1 (en) | System for, and method of, searching data records | |
| Xie et al. | Matching real-world facilities to building information modeling data using natural language processing | |
| US10127304B1 (en) | Analysis and visualization tool with combined processing of structured and unstructured service event data | |
| CN104750771B (en) | The method and system of context data analysis is carried out using domain information | |
| CN110188454B (en) | Building equipment and building information model matching method and device | |
| Delgado et al. | An evaluation of ontology matching techniques on geospatial ontologies | |
| Hu et al. | Knowledge extraction and discovery based on BIM: a critical review and future directions | |
| KR20180042710A (en) | Method and apparatus for managing a synonymous item based on analysis of similarity | |
| US20120239677A1 (en) | Collaborative knowledge management | |
| EP2633430A1 (en) | Generating a taxonomy from unstructured information | |
| CN113127506B (en) | Target query statement construction method and device, storage medium and electronic device | |
| Piasecki et al. | WordNetLoom: a WordNet development system integrating form-based and graph-based perspectives | |
| Amiridze et al. | Anti-unification and natural language processing | |
| Beneventano et al. | Semantic annotation of the CEREALAB database by the AGROVOC linked dataset | |
| CN111738008B (en) | Entity identification method, device and equipment based on multilayer model and storage medium | |
| KR102153259B1 (en) | Data domain recommendation method and method for constructing integrated data repository management system using recommended domain | |
| Mukherjee et al. | Automated knowledge provider system with natural language query processing | |
| Kefalidis et al. | Benchmarking geospatial question answering engines using the dataset GeoQuestions1089 | |
| WO2023103815A1 (en) | Contextual dialogue framework over dynamic tables |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230426 Address after: 100041 1008, building 16, xishanhui, Shijingshan Park, Zhongguancun Science Park, Haidian District, Beijing Applicant after: BIM WINNER (BEIJING) TECHNOLOGY CO.,LTD. Applicant after: BIM WINNER (SHANGHAI) TECHNOLOGY Co.,Ltd. Applicant after: SHENZHEN BIM WINNER TECHNOLOGY Co.,Ltd. Applicant after: Yingjia Internet (Beijing) Smart Technology Co.,Ltd. Address before: 100041 1008, building 16, xishanhui, Shijingshan Park, Zhongguancun Science Park, Haidian District, Beijing Applicant before: BIM WINNER (BEIJING) TECHNOLOGY CO.,LTD. Applicant before: BIM WINNER (SHANGHAI) TECHNOLOGY Co.,Ltd. Applicant before: JIAXING WUZHEN YINGJIA QIANZHEN TECHNOLOGY Co.,Ltd. Applicant before: SHENZHEN QIANHAI YINGJIA DATA SERVICE Co.,Ltd. Applicant before: SHENZHEN BIM WINNER TECHNOLOGY Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231213 Address after: 1308, 13th Floor, Shougang Sports Building, No. 6 Jinyuanzhuang Road, Shijingshan District, Beijing, 100043 Patentee after: BIM WINNER (BEIJING) TECHNOLOGY CO.,LTD. Address before: 100041 1008, building 16, xishanhui, Shijingshan Park, Zhongguancun Science Park, Haidian District, Beijing Patentee before: BIM WINNER (BEIJING) TECHNOLOGY CO.,LTD. Patentee before: BIM WINNER (SHANGHAI) TECHNOLOGY Co.,Ltd. Patentee before: SHENZHEN BIM WINNER TECHNOLOGY Co.,Ltd. Patentee before: Yingjia Internet (Beijing) Smart Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right |