Search

CN-115168392-B - Structural rule description method and system for building specification index query

CN115168392BCN 115168392 BCN115168392 BCN 115168392BCN-115168392-B

Abstract

The invention relates to a structural rule description method and a structural rule description system for building standard index query, which are characterized by comprising the steps of obtaining a multi-building BIM model of a to-be-queried item and corresponding SNL rules thereof, extracting entity data of the multi-building BIM model, determining corresponding entity relations, constructing a multi-building semantic model of the to-be-queried item, analyzing and obtaining a grammar structure of the SNL rules corresponding to the multi-building BIM model of the to-be-queried item, obtaining SNL for building index query, determining SPARQL query sentences for building indexes, querying the multi-building semantic model by adopting the SPARQL query sentences for the building indexes to obtain building standard indexes of the multi-building BIM model of the to-be-queried item, analyzing conclusion clauses for the multi-building BIM model for building index query, judging whether the conclusion clauses contain calculation or not, and outputting the corresponding building standard indexes as target building standard indexes according to judgment results.

Inventors

  • ZHANG HEHUA
  • YANG YIFEI

Assignees

  • 清华大学

Dates

Publication Date
20260505
Application Date
20220707

Claims (6)

  1. 1. The structural rule description method for building standard index query is characterized by comprising the following steps of: acquiring a multi-building BIM model of an item to be queried and an SNL rule corresponding to the multi-building BIM model; Extracting entity data of a multi-building BIM model, determining corresponding entity relations, and constructing a multi-building semantic model of the item to be queried; analyzing and acquiring a grammar structure of an SNL rule corresponding to a multi-building BIM model of an item to be queried, acquiring SNL for building index query, and determining SPARQL query sentences for building indexes; Adopting SPARQL query sentences oriented to building indexes in the multi-building semantic model to query so as to obtain building standard indexes of the multi-building BIM model of the item to be queried; Adopting an SNL grammar analysis framework to analyze an SNL conclusion clause facing to the building index query, judging whether the conclusion clause contains calculation or not, and outputting a corresponding building specification index as a target building specification index according to a judgment result, wherein if the conclusion clause contains calculation, the queried attribute information is backfilled into the corresponding attribute of the SNL in a left-to-right sequence, the corresponding attribute value is calculated in a middle-order calculation mode, and the calculation result is output as the target building specification index; the extracting entity data of the multi-building BIM model, determining corresponding entity relations, and constructing a multi-building semantic model of the item to be queried, comprising: Acquiring attribute information in a multi-building BIM model of an item to be queried, and storing the attribute information as an entity of a semantic model, wherein the multi-building BIM model comprises a plurality of single building models and a site model; Extracting entity data of a multi-building BIM model of the item to be queried layer by layer based on a macroscopic to microscopic mode, determining corresponding entity relations, and constructing a multi-building semantic model of the item to be queried; Analyzing and acquiring a grammar structure of an SNL rule corresponding to a multi-building BIM model of an item to be queried, acquiring SNL facing building index query, and determining SPARQL query sentences facing building indexes, wherein the method comprises the following steps: Analyzing and acquiring a grammar structure of an SNL rule corresponding to a multi-building BIM model of an item to be queried by adopting an SNL grammar analysis framework, and acquiring SNL facing to building index query; According to the SNL facing the building index query, obtaining a SPARQL query statement facing the building index, which comprises the following steps: Identifying target attributes of the multi-building BIM model of the item to be queried through the identifier; Identifying target attributes of the multi-building BIM model of the item to be queried through attribute mapping configuration; Skipping the generation of the corresponding SPARQL clauses of the conclusion clauses of the SNL facing the building index query, and placing the identifiers of the corresponding target attributes into the SELECT clauses according to the corresponding sequence in the SNL to obtain the SPARQL query statement facing the building index.
  2. 2. The method for describing the structural rules for building specification index query according to claim 1, further comprising: and counting the obtained target building standard index, and storing the target building standard index according to the format of 'index name-entity name-index value'.
  3. 3. The method for describing the structural rule for building specification index query according to claim 1, wherein the step of extracting the entity data of the multi-building BIM model of the item to be queried layer by layer based on a macroscopic to microscopic mode and determining the corresponding entity relationship to construct the multi-building semantic model of the item to be queried comprises the following steps: Extracting a single building model entity and a site model entity in a multi-building BIM model of an item to be queried, and storing the single building model entity and the site model entity in a corresponding entity inclusion relation; Extracting and organizing the entities in the single building model of the multi-building BIM of the item to be queried one by one, and storing the entities in the corresponding semantic model structure; And traversing all components in the multi-building BIM model of the item to be queried, and completing the construction of the multi-building semantic model.
  4. 4. A structural rule description system for building code index query, comprising: The input module is used for inputting a multi-building BIM model of the item to be queried and corresponding SNL rules; The semantic model construction module is used for extracting entity data of the multi-building BIM model, determining corresponding entity relations and constructing a multi-building semantic model of the item to be queried; The SNL expansion module is used for analyzing and acquiring a grammar structure of an SNL rule corresponding to the multi-building BIM model of the item to be queried, acquiring SNL facing the building index query, and determining SPARQL query sentences facing the building index; The query module is used for querying by adopting a SPARQL query statement oriented to the building index in the multi-building semantic model to obtain the building standard index of the multi-building BIM model of the item to be queried; The system comprises a conclusion clause analysis module, a construction index query module and a construction index query module, wherein the conclusion clause analysis module is used for analyzing a conclusion clause of an SNL facing the construction index query by adopting an SNL grammar analysis framework, judging whether the conclusion clause contains calculation or not, and outputting a corresponding construction specification index as a target construction specification index according to a judgment result, wherein if the conclusion clause contains calculation, the queried attribute information is backfilled into the corresponding attribute of the SNL in a left-to-right sequence, the corresponding attribute value is calculated in a middle-sequence calculation mode, and the calculation result is output as the target construction specification index; the extracting entity data of the multi-building BIM model, determining corresponding entity relations, and constructing a multi-building semantic model of the item to be queried, comprising: Acquiring attribute information in a multi-building BIM model of an item to be queried, and storing the attribute information as an entity of a semantic model, wherein the multi-building BIM model comprises a plurality of single building models and a site model; Extracting entity data of a multi-building BIM model of the item to be queried layer by layer based on a macroscopic to microscopic mode, determining corresponding entity relations, and constructing a multi-building semantic model of the item to be queried; Analyzing and acquiring a grammar structure of an SNL rule corresponding to a multi-building BIM model of an item to be queried, acquiring SNL facing building index query, and determining SPARQL query sentences facing building indexes, wherein the method comprises the following steps: Analyzing and acquiring a grammar structure of an SNL rule corresponding to a multi-building BIM model of an item to be queried by adopting an SNL grammar analysis framework, and acquiring SNL facing to building index query; According to the SNL facing the building index query, obtaining a SPARQL query statement facing the building index, which comprises the following steps: Identifying target attributes of the multi-building BIM model of the item to be queried through the identifier; Identifying target attributes of the multi-building BIM model of the item to be queried through attribute mapping configuration; Skipping the generation of the corresponding SPARQL clauses of the conclusion clauses of the SNL facing the building index query, and placing the identifiers of the corresponding target attributes into the SELECT clauses according to the corresponding sequence in the SNL to obtain the SPARQL query statement facing the building index.
  5. 5. A processing device comprising computer program instructions, wherein the computer program instructions, when executed by the processing device, are adapted to implement the steps corresponding to the building code index query-oriented structuring rule description method of any one of claims 1-3.
  6. 6. A computer readable storage medium, wherein computer program instructions are stored on the computer readable storage medium, wherein the computer program instructions, when executed by a processor, are configured to implement the steps corresponding to the building specification index query oriented structuring rule description method of any one of claims 1-3.

Description

Structural rule description method and system for building specification index query Technical Field The invention relates to the field of automatic detection, in particular to a structural rule description method and a structural rule description system for building standard index query. Background With rapid development and wide application of building information models, automatic inspection of building specifications is also a focus of attention of related industries. The prior art discloses a complex building code compliance detection device combining semantic and geometric calculations, however, the device can only realize compliance checking of building codes. Statistics and examination of various building related indexes are the important focused contents of examination works such as building planning report and construction, and the output result of the index examination work is usually a specific index value, but not a "correct" or "wrong" compliance result. In order to obtain and count various building related indexes, an effective data organization and extraction method is required. In addition, since building planning indexes generally adopt local standard standards, the definition mode of the building planning indexes is related to the geographical position of the project, the type of the project and other factors. Therefore, the extensible building index description and definition method is also significant for realizing efficient building index extraction and query. At present, the existing enterprises develop an automatic inspection tool for building planning report and construction aiming at a building information model, and the automatic inspection tool comprises the functions of statistics and derivation of building planning report and construction indexes. The tool enables marking of different types of components, such as green, parking spaces, etc., through a Revit-aided modeling plug-in. The tool counts and generates a report of the related indexes through a preset form, and the form of the form is shown in figure 1. The tool realizes statistics of common building planning indexes by combining auxiliary modeling with a preset statistical algorithm. However, the statistical method is invisible to the user, so that the user cannot realize the query of the target index in a mode of customizing the statistical rule. In addition, distinguishing component categories by tagging will also add additional modeling effort to the user. Currently, many automated specification review tools are similar to the tools described above, i.e., there are highly customized, lack of scalability issues. The BIMChecker model checking tool developed by the university of Qinghua software college solves the problem, and the tool defines and describes checking rules by using structured natural language (Structured Natural Language, SNL) and processes the checking rules by combining with a corresponding checking engine to realize checking of the self-defined rules, thereby realizing expansion of building specifications of different fields and different types. The tool generates the BIM model into the corresponding semantic model to realize the check of semantic information, however, as the tool only supports the generation of the semantic model of the single building, the tool can only perform standard check on the interior of the single building, but cannot support building related standard check under larger scale, such as building planning report check on multiple building models. Disclosure of Invention Aiming at the problems, the invention aims to provide a structural rule description method and a structural rule description system for inquiring building standard indexes, which can report, build and examine building planning of multiple building models. In order to achieve the aim, the invention adopts the following technical scheme that in the first aspect, a structural rule description method facing to building standard index query is provided, which comprises the following steps: acquiring a multi-building BIM model of an item to be queried and an SNL rule corresponding to the multi-building BIM model; Extracting entity data of a multi-building BIM model, determining corresponding entity relations, and constructing a multi-building semantic model of the item to be queried; analyzing and acquiring a grammar structure of an SNL rule corresponding to a multi-building BIM model of an item to be queried, acquiring SNL for building index query, and determining SPARQL query sentences for building indexes; Adopting SPARQL query sentences oriented to building indexes in the multi-building semantic model to query so as to obtain building standard indexes of the multi-building BIM model of the item to be queried; And analyzing a conclusion clause of the SNL facing the building index query, judging whether the conclusion clause contains calculation, and outputting a corresponding building standard index as a target bui