CN110134724A - A kind of the data intelligence extraction and display system and method for Building Information Model - Google Patents
A kind of the data intelligence extraction and display system and method for Building Information Model Download PDFInfo
- Publication number
- CN110134724A CN110134724A CN201910401139.XA CN201910401139A CN110134724A CN 110134724 A CN110134724 A CN 110134724A CN 201910401139 A CN201910401139 A CN 201910401139A CN 110134724 A CN110134724 A CN 110134724A
- Authority
- CN
- China
- Prior art keywords
- data
- unit
- bim
- user
- entity
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000000605 extraction Methods 0.000 title claims description 8
- 238000003860 storage Methods 0.000 claims abstract description 42
- 238000013500 data storage Methods 0.000 claims abstract description 18
- 238000009826 distribution Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000013075 data extraction Methods 0.000 claims abstract description 4
- 239000000284 extract Substances 0.000 claims description 14
- 238000012805 post-processing Methods 0.000 claims description 14
- 238000010276 construction Methods 0.000 claims description 12
- 230000003044 adaptive effect Effects 0.000 claims description 10
- 230000000694 effects Effects 0.000 claims description 10
- 238000013461 design Methods 0.000 claims description 9
- 238000007726 management method Methods 0.000 claims description 9
- 210000000352 storage cell Anatomy 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 210000004027 cell Anatomy 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 238000004040 coloring Methods 0.000 claims description 2
- 230000003993 interaction Effects 0.000 claims description 2
- 238000003672 processing method Methods 0.000 claims description 2
- 239000004744 fabric Substances 0.000 claims 1
- 238000002620 method output Methods 0.000 claims 1
- 230000006870 function Effects 0.000 description 12
- 238000004458 analytical method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000013523 data management Methods 0.000 description 4
- 238000012800 visualization Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 150000001875 compounds Chemical class 0.000 description 3
- 238000013079 data visualisation Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000002452 interceptive effect Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
Abstract
The invention discloses a kind of data intelligences for the Building Information Model for belonging to Internet technical field to extract and display system and method.A kind of method for specifically easily obtaining for layman and being provided using the magnanimity project data based on Building Information Model BIM.It includes: that magnanimity project data storage unit based on BIM and cloud, distribution BIM data pre-processing unit, user's natural language input and understand unit, domain knowledge storage unit, query statement generation unit, query execution unit, Data Post unit and self-adapting data display unit that the data intelligence of the Building Information Model that the present invention redesigns, which is extracted with display system,.The present invention calculates the intuitive displaying that data are realized with the mutual linkage of table, datagram, threedimensional model using natural language understanding, intelligent Data Extraction Based, it improves layman to extract from magnanimity BIM data, understand the efficiency in relation to data, reduces learning cost and operation difficulty.
Description
Technical field
The invention belongs to Internet technical field, in particular to the data intelligence of a kind of Building Information Model is extracted and display
System and method.It specifically easily obtains and is utilized based on Building Information Model (building for layman
Information model, BIM) magnanimity project data, provide based on a kind of BIM data intelligence by natural language extracts
Calculation and display system and method,
Background technique
Building Information Model is to be integrated with the work of the various relevant informations of construction-engineering project based on 3-dimensional digital technology
Journey data model is the digital expression to engineering project infrastructure entities and functional characteristic.One perfect information model, can
Data, process and the engineering resource of connecting building project life cycle different phase are the complete descriptions to engineering object, can be built
If project owner, design side, construction party and user etc. generally use.BIM is the consolidated project data in construction project each stage
Source completely integrate the Various types of data such as engineering design, construction and text, picture, the monitoring of each stage design of operation, and supports each ginseng
With square inquiry during the work time, utilization, update and storage information, ancillary works management and decision.
IFC (industry foundation classes) standard by international co-operation alliance in 1997 and by
The building engineering field information model standard of buildingSMART organizational protection.IFC standard uses EXPRESS language definition
Incidence relation between all kinds of related entities and entity of architectural engineering provides architectural engineering towards building full lifetime
Data representation and storage standard are that presently most comprehensive, the highest BIM standard of acceptance level and each BIM software support are best
Standard, it has also become architecture information describes the fact that field property standard.For all letters for completely describing construction project full lifetime
Breath, IFC standard content is very huge, structure is also considerably complicated, and only key message entity just has as many as several hundred, gives professional
Normal structure, content is understood completely, flexibly brings extreme difficulties using related data.
Since construction project has the characteristics that the scale of construction is huge, the period is long, profession is numerous, each stage is designed, constructs and runed
Mass data is used and produced, also relates to each professional domain such as progress, cost, quality, each stage, field are often
It using different software, data format and interdepends, therefore, construction project data have typical big data feature.In addition,
Each specialized service of construction project all relies on other expert datas, and engineering management and operation stage further relate to the comprehensive of multiple expert datas
Analysis and utilization is closed, correspondingly, related engineering staff is also required to grasp multi-specialized knowledge, to the generation of project data, manages and makes
With proposing very high requirement.
Currently, mainly the project data based on BIM is managed and is inquired, utilized by following three kinds of modes:
(1) mode based on existing BIM software and its internal data format
Which is based on existing business BIM software, using the internal data format storage and management BIM of corresponding software
Data, Object Selection, the data that inquiry and acquisition in relation to data can be provided by software are checked or export function is realized.Such as base
Major design stage BIM information can be stored in Autodesk Revit software and corresponding rvt format, and Autodesk can be passed through
The component selection of Revit, structure attribute check, the functional inquiries such as detail list, browsing for information about and threedimensional model.Which is wanted
It asks user of service to have more deep understanding to BIM software, and knows using which function the inquiry for realizing corresponding data and obtain
It takes.
(2) mode based on relevant database and its query language
Which utilizes data with the relevant databases management and construction engineering BIM data such as mysql, SQL Server
The modes such as the SQL query language of offer carry out data query and acquisition.Which needs computer professional voluntarily to build to magnanimity
If project data and its structure carry out relational data modeling, and it is deeper to require related engineering staff to have SQL query language
Solution, and voluntarily write inquiry and acquisition that query statement realizes data.Corresponding data result usually in a tabular form based on, need from
Row carries out visualization processing to data using other software.
(3) mode of the BIM data management platform based on database
Which is further to develop what data management platform obtained on the basis of aforesaid way (2).The data of exploitation
Management platform usually provides data importing and management function, and according to specific transactions Scenario Design and can realize that corresponding data are in
Now and visualization interface.Which has embedded query sentence of database in platform, and it is corresponding with interface function, makes user
It need not be concerned about that background data base is realized and query statement writes problem.Related engineering staff needs learning data first to manage platform
Application method, need in practical application that corresponding function is selected to realize data query and acquisition according to its business datum demand, and
Data presentation is carried out using the display mode of platform default.
With the popularization and application of the technologies such as Internet of Things, mobile terminal, the construction project data scale based on BIM will be increasingly huge
Greatly, how to quickly find that need information will be severe challenge that related engineering staff faces from magnanimity project data.However,
That there are still interactive modes is unfriendly for prior art method, flexibility ratio is low, relies on professional knowledge asks so that data display effect is general etc.
Topic is made a concrete analysis of as follows:
(1) mode based on existing BIM software: current, existing BIM software is mainly with design and construction management software
Main, interface and function all refer to multiple professions, and function is more, interface layout is complicated, and related engineering staff needs the long period to learn
Habit could grasp specific software usage mode.Meanwhile related data are shown and model visualization function is based on specific business,
It is in terms of data acquisition, displaying and not flexible, or need a large amount of secondary development that can meet data acquisition demand.Therefore,
For non-BIM professional, needs the long period that could grasp related software and use, and be not well positioned to meet its data and obtain
Take and present demand.For example, Autodesk Revit is related to building, structure, electromechanical three big profession at present, each profession has
Tens, a function even up to a hundred, learning difficulty is big, the time is long, is difficult for non-professional administrative staff or owner from it
Middle quick obtaining for information about, and data presentation be also from the angle of designer present, cannot show from visual angle
Data relationship is not able to satisfy the demand of owner or administrative staff.
(2) mode based on relevant database and its query language: the database creation of which and query statement are compiled
It writes and is required to related business personnel database design and its query language are well understood by, therefore learning difficulty is very big, to common
It is very unfriendly for engineering staff.Meanwhile data export mostly in the form of table etc., need to be transformed into other software or tool carries out
Data preparation and visual work, it is time-consuming and laborious.
(3) mode of the BIM data management platform based on database: which increases interactive mode in Basis of Database
Interface and data query function.But the relevant personnel, which need certain time to give training, can grasp platform feature.Meanwhile platform is only
It can be in a manner of being set in advance and function carries out data acquisition and displaying, it then needs further to develop if you need to new data exhibition method
And extension.Its data visualization effect is also limited by concrete function, and the mutual of data, model and document can not be presented well
Association, for owner and administrative staff, it is difficult to effectively see the complete of data
For interactive mode existing for above-mentioned current BIM data management, acquisition and display technology is complicated, flexibility ratio is low, according to
The problem of relying professional knowledge, data display effect general and lacking data correlation, is originally researched and proposed a kind of based on natural language
BIM data intelligence extracts calculating and display system and method, allows the unprofessional users such as administrative staff, owner can be by natural language
Speech expression data acquisition demand, by computer management magnanimity BIM data and automatic understanding user demand, the intelligence of data needed for realizing
It can extract, calculating shows with dynamic and visual.
Summary of the invention
The object of the present invention is to provide a kind of data intelligences of Building Information Model to extract and display system and method, spy
Sign is, the data intelligence of the Building Information Model of redesign extract and display system by BIM and cloud magnanimity project data
Storage unit, distribution BIM data pre-processing unit, the input of user's natural language understand unit, domain knowledge storage unit, look into
Ask sentence generation unit, query execution unit, Data Post unit and self-adapting data display unit composition;Wherein BIM and
The magnanimity project data storage unit of cloud is bi-directionally connected with distributed BIM data pre-processing unit and query execution unit respectively,
Query execution unit is connect with query statement generation unit and data post-processing unit respectively again, and the connection of Data Post unit is certainly
Adapt to data display unit;The input of user's natural language understands unit, domain knowledge storage unit and query statement generation unit
It is sequentially connected in series into circuit;With being described as follows:
The magnanimity project data storage unit of the BIM and cloud: the unit is based on IFC standard, with IFC storage method
The efficient storage and management for realizing BIM data database beyond the clouds, can carry PB grade project data, and data storage and updating is imitated
Rate promotes 4-6 times;
The distribution BIM data pre-processing unit: the unit uses MapReduce distributed data processing method, adopts
With distributed BIM data pre-connection method, realize that the associated high speed precomputation of IFC entity complex and cloud caching, promotion are looked into
Ask 1.5-1.8 times of efficiency;
User's natural language input understands unit: intention intelligent Understanding of the unit based on user automatically analyzes use
The natural language at family inputs, and the kernel entity and its binding characteristic of identification user's input support subsequent intelligent Data Extraction Based unit
Realization;
The domain knowledge storage unit: the unit is with ontology library or other knowledge base forms storage building trade in relation to knowing
Know, specifically includes each key concept object of IFC standard and its attribute definition, correlation, industry slang synonym and usual word
Information, to be intended to the user to be converted to Unified Form and establish and the relationship of IFC standard lays the foundation;
The query statement generation unit: the unit is based on the input of user's natural language and understands unit, domain knowledge storage
The output of unit, according to specific cloud database data query mode, automatically generated data query script or querying command;
The query execution unit: the unit executes the query script or inquiry that query statement generation unit exports for automatic
Order extracts the data that user is concerned about from cloud database, and by the conceptual object and its Attribute Association of each data and knowledge base
Get up, is output to Data Post unit;
The Data Post unit: the unit is intended to understand that result carries out certainly the output of query execution unit according to user
Dynamic post-processing, including classified, summarized or calculated according to a certain feature;And result is output to unit 8;
The adaptive BIM data display unit: according to the output of Data Post unit as a result, automatically selecting suitably
Data display mode shows hidden, coloring, the control of scaling including curve graph, histogram, table and three bit models;Using image,
The mode that data, model are interrelated, interact displays data;Open-and-shut displaying is in front of the user.
The data intelligence of the Building Information Model extracts the architecture information intelligent extraction and display methods with display system,
Characterized by comprising the following steps:
(1) magnanimity BIM data store: utilizing the BIM data distribution formula storage method based on IFC and cloud, design database
Storage strategy and each tables of data construct data storage cell.Meanwhile construction project BIM data are imported into number using IFC interface
According to storage unit 1, the storage of magnanimity BIM data is realized;
(2) data pre-association is calculated and is cached: the data pre-association calculation method based on MapReduce is utilized, in conjunction with IFC
The correlation rule of each entity extracts each data table data of data storage cell, carries out the calculating of data pre-association, and will finally close in advance
Join calculated result storage or update and arrives data storage cell.
(3) understand that user's natural language inputs: by providing text retrieval frame, unprofessional user being allowed to input with natural language
Data query requirements;Then, user is inputted using natural language input intelligent Understanding method and is converted into syntax tree structure, and from
Domain knowledge storage unit query relevant information establishes the incidence relation of syntax tree and domain knowledge, IFC entity;
(4) it generates data query sentence: on the basis of user's natural language inputs and understands, utilizing the BIM number of step (3)
It is investigated that asking sentence generation method and converting syntax tree structure for user's input, domain knowledge memory cell data is read, is passed through
Knowledge base graph search establishes the associated path of each IFC entity, and automatically generates query statement;
(5) it runs query statement: automatically generating query statement and query execution unit, operation inquiry language using step (4)
Sentence extracts related data from data storage cell, and is transferred to follow-up data post-processing unit.
(6) carry out Data Post: notebook data post-processing unit is grouped query execution cell data, sorts out and converges
It is total to carry out calculation processing, and it is transferred to data display unit;
(7) data are presented to show: utilize adaptive BIM data display unit, the data of step (6) post-processing are subjected to table
The mode that lattice, datagram are combined with threedimensional model is presented, and intuitively shows table, datagram and three by display effect
The correlativity of dimension module understands convenient for user and establishes its connection.
BIM data distribution formula storage method based on IFC and cloud in the step (1) be IFC entity is divided into O, RL, P,
G, RLx totally 5 class, in which: O represents all entities inherited from IfcObjectDefinition, RL represent it is all from
The entity that IfcRelationship is inherited, P represent all entities defined in the resource layer of IFC outline and data type, G generation
All entities for indicating geometry of Table I FC standard, RLx is represented included in O and P but the entity as relationship object;
The storage strategy and method of above-mentioned 5 class entity are as follows: the entity that O class, RL and RLx class include stores respectively to be independent
Tables of data;And P class entity is then used as O, RL and RLx class entity attributes to be stored in corresponding tables of data, does not store individually;G
Class entity need to use different storage methods according to type of database and query demand, work as database data because data volume is larger
When entity attributes allow size of data biggish, it can be stored G class entity as O class entity attributes, it in this way can be with
It is time-consuming to avoid correlation inquiry when inquiry geological information, and when the spatial relationship or database data entity for needing consideration complicated
When attribute allows size of data smaller, Ying Jiang G class entity is separately stored as independent tables of data.
Step (6) concrete operations, data are grouped, reconstruct and are summarized using inquiry data post processing method,
It calculates, to realize subsequent adaptive display effect, the specific steps are as follows:
1) user is read to be intended to understand result: by above-mentioned user be intended to the output of intelligent Understanding method crucial name entity,
Related domain knowledge entity and IFC entity are read in;
2) association user intention understands result: above-mentioned reading data being associated with query result, to expand each
Inquire data semantic information, so as to computer understanding inquiry the data obtained belong to which IFC entity, attribute and they
How is correlation, to provide support for data grouping calculating;
3) data grouping and structural adjustment: according to inquiry data and associated semantic information, to same alike result value
Data be grouped, in combination with other semantic features building data hierarchical tree structure etc.;
4) data calculating summarizes: it is intended to understand according to user as a result, combined data semantic information,
Automatically the data after grouping are summarized, operation of summing, average etc., preferably to realize that data are aobvious
Show.
It is extracted the beneficial effects of the invention are as follows the data intelligence of the Building Information Model of the invention redesigned and is with display
The storage of data distribution formula, processing and the user that system realizes BIM are intended to understanding, intelligent Data Extraction Based calculates and table, data
The interaction association of figure, threedimensional model is shown with intuitive, is effectively increased layman and is extracted and understand from magnanimity BIM data
Efficiency in relation to data reduces learning cost and operation difficulty.The efficient storage of present invention support magnanimity architectural engineering data
With pretreatment, and it is appreciated that the data query requirements that user is inputted in the form of natural language, automatic correlation engineering data of realizing
The demand calculated with display data is extracted, layman can be substantially improved and obtain, utilize the efficiency of magnanimity architectural engineering data
Detailed description of the invention
Fig. 1 is system composition schematic diagram.
Fig. 2 is that the natural language of domain knowledge inputs intelligent Understanding schematic diagram.
Fig. 3 is the BIM query sentence of database product process figure based on graph search.
Fig. 4 is the Data Post schematic diagram that user oriented is intended to.
Fig. 5 is adaptive BIM data displaying schematic diagram.
Specific embodiment
The data intelligence that the present invention provides a kind of Building Information Model is extracted and display system and method, with reference to the accompanying drawing
And embodiment, system and implementation method that the present invention redesigns is discussed in detail:
Fig. 1 show system composition schematic diagram.The data intelligence of the Building Information Model of redesign shown in FIG. 1 is extracted
It include: magnanimity project data storage unit 1, distribution BIM data pre-processing unit 2, use based on BIM and cloud with display system
Natural language input in family understands unit 3., domain knowledge storage unit 4, query statement generation unit 5, query execution unit 6, number
According to post-processing unit 7 and self-adapting data display unit 8;
The intelligent extraction of architecture information of the present invention includes: with display methods
(1) the BIM data distribution formula storage method based on IFC and cloud
IFC standard describes BIM data entity using object-oriented method, and each entity is there are deep layer inheritance and has big
Attribute information and interrelated is measured, the tables of data that all kinds of entities are mapped to data directly can be generated into several hundred a tables of data, so that
Data query, extraction and incidence relation analysis efficiency are extremely low.Therefore, this patent Based on Distributed database proposes following storage
Strategy and method:
1) IFC entity is divided into O, RL, P, G, RLx totally 5 classes, in which: O represents all from IfcObjectDefinition
The entity of succession, RL represent all entities inherited from IfcRelationship, and it is fixed that P represents all resource layers in IFC outline
The entity and data type of justice, G represent all entities for indicating geometry of IFC standard, and RLx, which is represented, to be included in O and P but as pass
It is the entity of object.
2) storage strategy and method of above-mentioned 5 class entity are as follows: the entity that O class, RL and RLx class include is stored as independence respectively
Tables of data;And P class entity is then used as O, RL and RLx class entity attributes to be stored in corresponding tables of data, does not store individually;
G class entity need to use different storage methods according to type of database and query demand, work as database data because data volume is larger
When entity attributes allow size of data biggish, it can be stored G class entity as O class entity attributes, it in this way can be with
It is time-consuming to avoid correlation inquiry when inquiry geological information, and when the spatial relationship or database data entity for needing consideration complicated
When attribute allows size of data smaller, Ying Jiang G class entity is separately stored as independent tables of data.
Embodiment: it according to the method described above, is realized based on IFC standard 2x3 and 4.0 versions and MongoDB cloud database
The distributed storage of BIM data, actual test show that this method correlation traditional Relational DataBase is mentioned in data insertion speed
Rise 4 times, data renewal speed promotes 6 times and data query speed keeps fairly horizontal.
(2) the data correlation pre-computation methods based on MapReduce
As previously mentioned, the entity in IFC model is stored in different tables of data, and data can be dispersed to each collection
Group.Complicated data connection operation is needed in view of a large amount of IFC entities are there are interrelated, when cloud is inquired, exist it is time-consuming it is high,
The problem of low efficiency.Therefore, using the data correlation pre-computation methods based on MapReduce.To A class entity and B class entity,
Middle A class entity stores the key assignments BKey of B class entity, for the association precomputation for realizing the two;Following steps can be used:
1) according to association key assignments BKey, Map process is executed to A class physical data table and B class physical data table respectively;
2) based on association key assignments BKey, Reduce process is executed to the output result of above-mentioned steps, generates temporary data set
It closes;
3) key assignments is exported according to target, the ephemeral data set generated to above-mentioned steps executes Map process again;
4) the output result of the above process is stored to database, or database information is updated according to data result.
Embodiment: according to above-mentioned steps, based on javascript, to realize data pre- in MongoDB cloud database
It is associated with calculating process, for the collection of material of beam set and 10,000 records with 1,000,000 records, efficiency data query can be promoted
1.5 to 1.8 times.
(3) natural language based on domain knowledge inputs intelligent Understanding method
In order to which the natural language of intelligent Understanding unprofessional user inputs, present invention introduces industry knowledge bases, propose nature language
Speech input intelligent Understanding method, specifically includes the following steps:
1) text word segmentation processing: use hidden markov model or other machines learning method by user input cutting for
A series of words, while removing stop words, modal particle etc., the input as subsequent step;
2) part-of-speech tagging: carrying out part-of-speech tagging to the word segmented based on machine learning or statistical method, clear each
Word is noun, verb, numeral-classifier compound or preposition etc.;
3) syntactic analysis: according to part-of-speech tagging as a result, the syntactic structure of anolytic sentence, converts grammer for user's input
Tree;
4) entity classification: analysis syntax tree structure identifies crucial name entity by extreme saturation, breadth traversal algorithm
(such as beam, column, project amount), meanwhile, analysis numeral-classifier compound, preposition, conjunction etc. and their positions in syntax tree determine each
The constrained parameters information of key name entity;
5) industry knowledge base is associated with: being introduced industry knowledge base, by measuring similarity, is established crucial name entity and its about
Beam parameter is associated with industry knowledge base, and realizes that synonym, homophone are unified by industry knowledge base, numeral-classifier compound normalization,
User is avoided to express ambiguity.
Embodiment: being based on this method, and the present invention uses the natural language processing tool Stanford of Stanford University
NLP realizes text participle, part-of-speech tagging and syntactic analysis, and voluntarily realizes crucial name entity identification algorithms and know with industry
Know library association (as shown in Figure 2).
(4) the BIM data query sentence generation method based on graph search
It is specific real using the BIM query sentence of database generation method based on graph search based in the above way exporting
Apply that steps are as follows:
1) industry knowledge base is read, while producing the start-up portion of query statement;
2) the natural language understanding result of above step is read;
3) position of crucial the name entity and its parameter of identification in industry knowledge base;
4) it based on graph structures path search algorithms such as dijkstra, finds through the road in relation to name entity and its parameter
Diameter, and record all path entities and its correlation;
5) it since path originates entity, successively traverses backward;
6) to each entity, the corresponding query statement of the entity is obtained first and generates template, while obtaining the entity
Restriction on the parameters or querying condition, and by the template of corresponding information filling, ultimately produce the query statement segment of the entity;
7) after traversing, querying command termination message is generated.
Embodiment: being based on above step, and using MongoDB cloud database is based on, industry knowledge base is stored as figure
Structure type realizes route searching and querying method generates (as shown in Figure 3) using dijkstra algorithm.
(5) the inquiry data post processing method that user oriented is intended to
After obtaining data query result, data are grouped, reconstruct and are summarized, are counted using inquiry data post processing method
It calculates, to realize subsequent adaptive display effect, the specific steps are as follows:
1) user is read to be intended to understand result: by above-mentioned user be intended to the output of intelligent Understanding method crucial name entity,
Related domain knowledge entity and IFC entity are read in;
2) association user intention understands result: above-mentioned reading data being associated with query result, to expand each
Inquire data semantic information, so as to computer understanding inquiry the data obtained belong to which IFC entity, attribute and they
How is correlation, to provide support for data grouping calculating;
3) data grouping and structural adjustment: according to inquiry data and associated semantic information, to same alike result value
Data be grouped, in combination with other semantic features building data hierarchical tree structure etc.;
4) data calculating summarizes: it is intended to understand according to user as a result, combined data semantic information,
Automatically the data after grouping are summarized, operation of summing, average etc., preferably to realize that data are shown
(as shown in Figure 4).
(6) adaptive BIM data display method
Data Post based on above-mentioned steps, can be automatically according to inquiry as a result, the adaptive BIM data display method
The characteristics of data result, automatically selects suitable display mode, specific as follows:
1) output data table: it is different according to the data structure of query result, mode is created using different tables.Its
In, array or list type data directly generate two-dimensional table;The data of tree structure then use extensible tree list shape
Formula output;Reticular structure data are also exported using two-dimensional table form, and are increased each node in last column and directly mutually interconnected
Connect information.
2) drawing data figure: it is similar, data visualization is also carried out using different data figure according to the form of query result.
Wherein, array class data are using forms such as bar chart, histograms;Two-dimensional array and list can further use curve graph, scatterplot
The forms such as figure;Tree data are using the forms visualization such as classification pie chart, rising sun figure;Reticular structure directlys adopt network etc.
Form is shown;While data visualization, automatically according to data distribution interval selection ordinary coor or logarithmic coordinates, more added with
Imitate display data.
3) show threedimensional model: the relationship based on data and model, by unrelated model hide or translucentization, will be related
Model show and highlight, meanwhile, be based on visible model information, by scaling of model to suitable ratio.
4) unified figure, table, model display effect: according to the relationship of data and model, consolidated table data, datagram and three
Same class data are shown using same filling mode, realize the displaying of three's correlation by dimension module.
Embodiment: according to above-mentioned steps, the collaboration for realizing inquiry data form, atlas threedimensional model is shown, specifically
Step.Based on the present invention, user can input the project amount of tri-layer beams " two layers and " and obtain project amount datagram shown in fig. 5, table
Lattice and model are shown as a result, to allow unprofessional user more easily to obtain data, more intuitive understand data.
Claims (4)
1. a kind of data intelligence of Building Information Model is extracted and display system, which is characterized in that the architecture information of redesign
The data intelligence of model extract and display system by the magnanimity project data storage unit of BIM and cloud, distribution BIM data in advance
Reason unit, the input of user's natural language understand unit, domain knowledge storage unit, query statement generation unit, query execution list
Member, Data Post unit and self-adapting data display unit composition;The wherein magnanimity project data storage unit of BIM and cloud point
Be not bi-directionally connected with distributed BIM data pre-processing unit and query execution unit, query execution unit again respectively with inquiry language
Sentence generation unit is connected with data post-processing unit, and Data Post unit connects self-adapting data display unit;User is natural
Language in-put understands that unit, domain knowledge storage unit and query statement generation unit are sequentially connected in series into circuit;
With being described as follows:
The magnanimity project data storage unit of the BIM and cloud: the unit is based on IFC standard, with the realization of IFC storage method
The efficient storage and management of BIM data database beyond the clouds, can carry PB grades of project data, data storage and update efficiency and mention
Rise 4-6 times;
The distribution BIM data pre-processing unit: the unit use MapReduce distributed data processing method, using point
Cloth BIM data pre-connection method realizes the associated high speed precomputation of IFC entity complex and cloud caching, promotes inquiry effect
1.5-1.8 times of rate;
User's natural language input understands unit: the intelligent Understanding of intention of the unit based on user automatically analyzes user
Natural language input, identification user input kernel entity and its binding characteristic, support subsequent intelligent Data Extraction Based unit
It realizes;
The domain knowledge storage unit: the unit stores the related knowledge of building trade with ontology library or other knowledge base forms,
Specifically include each key concept object of IFC standard and its attribute definition, correlation, the letter of industry slang synonym and usual word
Breath, for user to be intended to be converted to Unified Form and establish to lay the foundation with the relationship of IFC standard;
The query statement generation unit: the unit is based on the input of user's natural language and understands unit, domain knowledge storage unit
Output, according to specific cloud database data query mode, automatically generated data query script or querying command;
The query execution unit: the query script or querying command that the unit exports automatic execution unit, from cloud data
The data that user is concerned about are extracted in library, and the conceptual object and its Attribute Association of each data and knowledge base are got up, and are output to unit;
The Data Post unit: the unit is intended to understand that result post-processes unit output automatically according to user, wraps
It includes and is classified, summarized or calculated according to a certain feature;And result is output to unit;
The adaptive BIM data display unit: according to the output of Data Post unit as a result, automatically selecting suitable data
Display mode shows hidden, coloring, the control of scaling including curve graph, histogram, table and three bit models;Using image, number
Interrelated according to, model, interaction mode displays data;Open-and-shut displaying is in front of the user.
2. a kind of data intelligence of Building Information Model extracts the architecture information intelligent extraction and display methods with display system,
It is characterized in that, comprising the following steps:
(1) magnanimity BIM data store: utilizing the BIM data distribution formula storage method based on IFC and cloud, design database storage
Tactful and each tables of data constructs data storage cell;Meanwhile construction project BIM data are imported into data using IFC interface and are deposited
Storage unit realizes the storage of magnanimity BIM data;
(2) data pre-association is calculated and is cached: the data pre-association calculation method based on MapReduce is utilized, in conjunction with each reality of IFC
The correlation rule of body extracts each data table data of data storage cell, carries out the calculating of data pre-association, and finally by pre-association meter
It calculates result storage or updates and arrive data storage cell;
(3) understand that user's natural language inputs: by providing text retrieval frame, allowing unprofessional user with natural language input data
Query demand;Then, syntax tree structure is converted by user's input using natural language input intelligent Understanding method, and from industry
Knowledge storing unit query-related information establishes the incidence relation of syntax tree and domain knowledge, IFC entity;
(4) it generates data query sentence: on the basis of user's natural language inputs and understands, being looked into using the BIM data of step (3)
It askes sentence generation method and converts syntax tree structure for user's input, read domain knowledge memory cell data, pass through knowledge
Library graph search establishes the associated path of each IFC entity, and automatically generates query statement;
(5) it runs query statement: automatically generating query statement and query execution unit using step (4), run query statement, from
Data storage cell extracts related data, and is transferred to follow-up data post-processing unit;
(6) carry out Data Post: notebook data post-processing unit query execution cell data is grouped, sort out and is summarized into
Row calculation processing, and it is transferred to data display unit;
(7) data are presented show: utilizing adaptive BIM data display unit, by the data of step (6) post-processing progress table,
The mode that datagram is combined with threedimensional model is presented, and intuitively shows table, datagram and three-dimensional by display effect
The correlativity of model understands convenient for user and establishes its connection.
3. the data intelligence of Building Information Model extracts the architecture information intelligent extraction with display system according to claim 2
With display methods, which is characterized in that the BIM data distribution formula storage method based on IFC and cloud in the step (1) is by IFC
Entity is divided into O, RL, P, G, RLx totally 5 class, in which: O represents all entities inherited from IfcObjectDefinition, RL generation
All entities inherited from IfcRelationship of table, P represent all entities defined in the resource layer of IFC outline and data
Type, G represent all entities for indicating geometry of IFC standard, and RLx is represented included in O and P but the entity as relationship object;
The storage strategy and method of above-mentioned 5 class entity are as follows: the entity that O class, RL and RLx class include stores respectively is independent data
Table;And P class entity is then used as O, RL and RLx class entity attributes to be stored in corresponding tables of data, does not store individually;G class is real
Body need to use different storage methods according to type of database and query demand, when database data entity because data volume is larger
Attribute when allowing size of data biggish, can be stored G class entity as O class entity attributes, in this way can be to avoid
Correlation inquiry when inquiring geological information is time-consuming, and when the spatial relationship or database data entity attributes for needing consideration complicated
When allowing size of data smaller, Ying Jiang G class entity is separately stored as independent tables of data.
4. the data intelligence of Building Information Model extracts the architecture information intelligent extraction with display system according to claim 2
With display methods, which is characterized in that step (6) concrete operations divide data using inquiry data post processing method
Group reconstructs and summarizes, calculates, to realize subsequent adaptive display effect, the specific steps are as follows:
1) it reads user to be intended to understand result: above-mentioned user is intended to the crucial name entity, related of intelligent Understanding method output
Domain knowledge entity and IFC entity read in;
2) association user intention understands result: above-mentioned reading data being associated with query result, to expand each inquiry
The semantic information of data, so that computer understanding inquiry the data obtained is to belong to which IFC entity, attribute and theirs is mutual
How is relationship, to provide support for data grouping calculating;
3) data grouping and structural adjustment: according to inquiry data and associated semantic information, to the number with same alike result value
According to being grouped, in combination with the hierarchical tree structure etc. of other semantic features building data;
4) data calculating summarizes: be intended to understand according to user as a result, combined data semantic information, automatically to the data after grouping into
Row summarizes, and sums and averages, preferably to realize that data are shown.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910401139.XA CN110134724A (en) | 2019-05-15 | 2019-05-15 | A kind of the data intelligence extraction and display system and method for Building Information Model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910401139.XA CN110134724A (en) | 2019-05-15 | 2019-05-15 | A kind of the data intelligence extraction and display system and method for Building Information Model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110134724A true CN110134724A (en) | 2019-08-16 |
Family
ID=67574044
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910401139.XA Pending CN110134724A (en) | 2019-05-15 | 2019-05-15 | A kind of the data intelligence extraction and display system and method for Building Information Model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110134724A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110889161A (en) * | 2019-12-11 | 2020-03-17 | 清华大学 | Three-dimensional display system and method for sound control building information model |
CN110968751A (en) * | 2019-11-29 | 2020-04-07 | 中国电力企业联合会电力建设技术经济咨询中心 | Dynamic display method and device for engineering project capital curve |
CN111178051A (en) * | 2019-12-30 | 2020-05-19 | 盈嘉互联(北京)科技有限公司 | Building information model self-adaptive Chinese word segmentation method and device |
CN112016142A (en) * | 2020-08-08 | 2020-12-01 | 重庆非科智地科技有限公司 | BIM model lightweight system, method and device |
CN112700529A (en) * | 2019-10-07 | 2021-04-23 | 奥比塞德技术公司 | Method and system for generating three-dimensional model according to standard document |
CN113158276A (en) * | 2020-12-21 | 2021-07-23 | 深圳嘉瑞建设信息科技有限公司 | Monitoring information visualization system based on BIM |
CN113641842A (en) * | 2021-08-11 | 2021-11-12 | 贵州百胜工程建设咨询有限公司 | BIM-based atlas query method |
CN114996361A (en) * | 2022-07-29 | 2022-09-02 | 阿里巴巴(中国)有限公司 | Data processing method and system |
CN116127960A (en) * | 2023-04-17 | 2023-05-16 | 广东粤港澳大湾区国家纳米科技创新研究院 | Information extraction method, information extraction device, storage medium and computer equipment |
CN116226037A (en) * | 2023-05-08 | 2023-06-06 | 山东同圆数字科技有限公司 | IFC format data detail dynamic extraction method and system based on Web |
CN116644084A (en) * | 2023-07-27 | 2023-08-25 | 上海秉匠信息科技有限公司 | Method, apparatus, device and storage medium for processing three-dimensional model member data |
CN117290379A (en) * | 2023-11-27 | 2023-12-26 | 华南理工大学 | System for converting natural language into SQL and Revit data interaction |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108416008A (en) * | 2018-02-28 | 2018-08-17 | 华南理工大学 | A kind of BIM product database semantic retrieving methods based on natural language processing |
CN108536724A (en) * | 2018-02-13 | 2018-09-14 | 西安理工大学 | Main body recognition methods in a kind of metro design code based on the double-deck hash index |
CN109344223A (en) * | 2018-09-18 | 2019-02-15 | 青岛理工大学 | A kind of Building Information Model management system and method based on cloud computing technology |
-
2019
- 2019-05-15 CN CN201910401139.XA patent/CN110134724A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108536724A (en) * | 2018-02-13 | 2018-09-14 | 西安理工大学 | Main body recognition methods in a kind of metro design code based on the double-deck hash index |
CN108416008A (en) * | 2018-02-28 | 2018-08-17 | 华南理工大学 | A kind of BIM product database semantic retrieving methods based on natural language processing |
CN109344223A (en) * | 2018-09-18 | 2019-02-15 | 青岛理工大学 | A kind of Building Information Model management system and method based on cloud computing technology |
Non-Patent Citations (1)
Title |
---|
JIA-RUI LIN 等: "A natural-language-based approach to intelligent data retrieval and representation for cloud BIM" * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112700529A (en) * | 2019-10-07 | 2021-04-23 | 奥比塞德技术公司 | Method and system for generating three-dimensional model according to standard document |
CN110968751A (en) * | 2019-11-29 | 2020-04-07 | 中国电力企业联合会电力建设技术经济咨询中心 | Dynamic display method and device for engineering project capital curve |
CN110889161A (en) * | 2019-12-11 | 2020-03-17 | 清华大学 | Three-dimensional display system and method for sound control building information model |
CN111178051A (en) * | 2019-12-30 | 2020-05-19 | 盈嘉互联(北京)科技有限公司 | Building information model self-adaptive Chinese word segmentation method and device |
CN112016142B (en) * | 2020-08-08 | 2024-01-30 | 重庆非科智地科技有限公司 | BIM model light weight system, method and device |
CN112016142A (en) * | 2020-08-08 | 2020-12-01 | 重庆非科智地科技有限公司 | BIM model lightweight system, method and device |
CN113158276A (en) * | 2020-12-21 | 2021-07-23 | 深圳嘉瑞建设信息科技有限公司 | Monitoring information visualization system based on BIM |
CN113641842A (en) * | 2021-08-11 | 2021-11-12 | 贵州百胜工程建设咨询有限公司 | BIM-based atlas query method |
CN114996361A (en) * | 2022-07-29 | 2022-09-02 | 阿里巴巴(中国)有限公司 | Data processing method and system |
CN116127960A (en) * | 2023-04-17 | 2023-05-16 | 广东粤港澳大湾区国家纳米科技创新研究院 | Information extraction method, information extraction device, storage medium and computer equipment |
CN116226037A (en) * | 2023-05-08 | 2023-06-06 | 山东同圆数字科技有限公司 | IFC format data detail dynamic extraction method and system based on Web |
CN116644084A (en) * | 2023-07-27 | 2023-08-25 | 上海秉匠信息科技有限公司 | Method, apparatus, device and storage medium for processing three-dimensional model member data |
CN116644084B (en) * | 2023-07-27 | 2023-09-26 | 上海秉匠信息科技有限公司 | Method, apparatus, device and storage medium for processing three-dimensional model member data |
CN117290379A (en) * | 2023-11-27 | 2023-12-26 | 华南理工大学 | System for converting natural language into SQL and Revit data interaction |
CN117290379B (en) * | 2023-11-27 | 2024-03-15 | 华南理工大学 | System for converting natural language into SQL and Revit data interaction |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110134724A (en) | A kind of the data intelligence extraction and display system and method for Building Information Model | |
CN112560275B (en) | Two-dimensional three-dimensional real-time linkage online drawing annotation checking system and method | |
CN103631882B (en) | Semantization service generation system and method based on graph mining technique | |
CN110599592B (en) | Three-dimensional indoor scene reconstruction method based on text | |
CN108509566B (en) | Network topology visualization method based on-cloud 95598 data release service system | |
CN106484808B (en) | A kind of traffic object holography electronic record Data Modeling Method | |
CN104317842A (en) | Intelligent data extraction and thematic map generation engine based on geographic information technology | |
CN110866123A (en) | Method for constructing data map based on data model and system for constructing data map | |
Lo Turco et al. | Data modeling for museum collections | |
CN110377751A (en) | Courseware intelligent generation method, device, computer equipment and storage medium | |
CN111078094B (en) | Distributed machine learning visualization device | |
Bonsma et al. | Handling huge and complex 3D geometries with Semantic Web technology | |
CN103678593B (en) | A kind of interactive space scene search method described based on spatial scene sketch | |
US20240095227A1 (en) | Chart engine | |
US10776351B2 (en) | Automatic core data service view generator | |
Ledur et al. | Towards a domain-specific language for geospatial data visualization maps with big data sets | |
CN115099315A (en) | Multi-source heterogeneous geographic information data semantic fusion conversion method based on CityGML | |
Liu | Component-driven pocedural modeling for ancient Chinese architecture of the Qing Dynasty | |
CN104317853B (en) | A kind of service cluster construction method based on Semantic Web | |
CN115983385B (en) | Space environment entity construction method and device | |
CN112565875A (en) | Method, device, equipment and computer readable storage medium for automatically generating video | |
CN115358477B (en) | Fight design random generation system and application thereof | |
Li et al. | Digital twin technology in intelligent manufacturing | |
Zhang et al. | A conceptual framework for the automated generalization of geological maps based on multiple agents and workflow | |
Hu et al. | Efficient procedural modelling of building Façades based on windows from sketches |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190816 |