CN-122022546-A - Automatic assessment method and system for construction state of building indoor member

CN122022546ACN 122022546 ACN122022546 ACN 122022546ACN-122022546-A

Abstract

The application discloses an automatic evaluation method and system for construction states of indoor components of a building, and relates to the technical field of building construction, wherein the method comprises the steps of obtaining indoor point cloud data of the building, preprocessing the point cloud data and generating a point cloud model; the method comprises the steps of identifying a column point cloud based on an AI algorithm, extracting a center point, generating an actual construction shaft network, executing a shaft network pair Ji Peizhun based on the actual construction shaft network and a design shaft network of a BIM model, extracting three-dimensional parameters of a member to be monitored from the registered BIM model, generating BIM bounding box parameters, calculating coverage, judging a construction state based on a quantization threshold, counting the duty ratio of members in each state according to a subentry project, calculating the progress deviation rate of an actual progress and a planned progress by combining the planned project amount and the planned project period in the pre-led BIM model, and generating a visual evaluation report. The application realizes the high efficiency and standardization of construction progress evaluation of the building indoor components.

Inventors

CAI WENHAO
CHEN YICAI
FANG SUCHANG
YAO RUIZHE
WANG WENQING
WANG FANGBIAO
HAN JIE
ZHAO YANHUI
Long Chengbiao

Assignees

中优云视(深圳)科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251229

Claims (9)

1. An automatic assessment method for the construction state of an indoor building component is characterized by comprising the following steps: acquiring indoor point cloud data of a building, preprocessing the point cloud data to remove interference points and generating a global point cloud model; Identifying column point clouds in the global point cloud model based on an AI algorithm, extracting center points, and generating an actual construction shaft network; executing a shaft network pair Ji Peizhun based on the actual construction shaft network and a design shaft network of a preset BIM model; extracting three-dimensional parameters of a member to be monitored from the registered BIM model, and generating BIM bounding box parameters based on the three-dimensional parameters; calculating an overlapping region of the global point cloud model and BIM bounding box parameters to obtain coverage data, calling a database which is pre-stored with a corresponding relation between a coverage threshold value and a component construction state, and searching the database based on the coverage data to obtain a matched component construction state; And counting the duty ratio of each state component according to the itemized engineering, combining the planned engineering quantity and the planned construction period in the pre-led BIM model, calculating the progress deviation rate of the actual progress and the planned progress, and generating a visual evaluation report.
2. The method for automatically evaluating the construction state of the building indoor component according to claim 1, wherein the preprocessing comprises the steps of removing interference points through a statistical filtering algorithm, preserving geometric features through voxel grid downsampling, and generating a globally consistent indoor point cloud model through a multi-station stitching algorithm based on overlapping point clouds.
3. The automated construction state evaluation method of building indoor components according to claim 1, wherein the AI algorithm comprises automatically identifying a pillar point cloud area from a global point cloud model based on a pillar point cloud identification algorithm of PointCNN improvement models, performing cylinder fitting on the identified pillar point cloud, and calculating cylinder center coordinates as a pillar center point.
4. The method for automatically evaluating the construction state of the building indoor member according to claim 1, wherein the generation of the actual construction shaft network is characterized by comprising the steps of connecting adjacent column center points in pairs to generate longitudinal and transverse axes, selecting column center points around the building with the farthest span, connecting the column center points to form boundary axes, and constructing the complete actual construction shaft network.
5. The method for automated evaluation of construction conditions of building interior components according to claim 1, wherein the shaft network is aligned and registered, and the method comprises calculating an optimal transformation matrix of an actual construction shaft network and a design shaft network of the BIM model by a least square method, synchronously calculating a distance difference between an actual axis and a design axis, and defining the distance difference as axis deviation data.
6. The method for automatically evaluating the construction state of an indoor building component according to claim 1, wherein the step of acquiring the indoor point cloud data comprises the steps of: Setting a laser scanner (1) to scan an indoor construction area to obtain point cloud data; A supporting mechanism (2) for supporting and moving the laser scanner (1) is arranged, and a guiding module (3) for guiding an operator to move according to a scanning path is arranged on the supporting mechanism (2); The supporting mechanism (2) comprises a supporting plate (21), a support (22) and a movable base (23), wherein the movable base (23) is arranged at the lower end of the support (22) and used as a movable foundation, the supporting plate (21) is fixedly connected to the upper end of the support (22), and the upper surface of the supporting plate (21) is horizontally arranged and used for being detachably connected with the bottom of the laser scanner (1).
7. The automated construction state evaluation method for building indoor components according to claim 6, wherein the guiding module (3) comprises a man-machine interaction unit and a control unit, the control unit is in data connection with a control system of the laser scanner (1), and the control unit is configured to: acquiring an indoor member and an electromechanical pipeline layout diagram in a preset BIM model; Generating a scanning path and a focus scanning attention item based on a layout and a preset scanning rule, wherein the scanning rule comprises scanning along the outside Zhou Shun of a room to obtain boundary data, and then performing focus scanning on key components such as a beam column, an electromechanical pipeline and the like; And according to the scanning path and the focus scanning attention item, the man-machine interaction unit is instructed to conduct guiding display.
8. The automated construction state evaluation method for building indoor components according to claim 7, wherein the method for acquiring the building indoor point cloud data further comprises the steps of arranging a marking mechanism (4) for performing jet marking on the ground along a scanning path on a supporting mechanism (2), wherein the marking mechanism (4) is electrically connected with a control unit, the control unit is electrically connected with a camera, and the acquisition range of the camera covers an indoor construction area; the marking mechanism (4) is configured to: spraying marks on the ground at intervals of a preset time length t; The control unit is configured to: If a scanning completion instruction initiated by a user is received, invoking image data fed back by a camera; Identifying the mark points based on the image data to obtain mark point distribution data; Based on the layout diagram, identifying positions of beam columns and electromechanical pipelines in the layout diagram to obtain key component distribution data; and comparing the mark point distribution data with the key component distribution data, evaluating the scanning quality of the key component according to the analysis result, and outputting a matching prompt if the evaluation result shows that the scanning quality condition does not accord with the preset scanning quality condition.
9. An automated assessment system for construction status of building indoor components, comprising: the point cloud data acquisition module is used for acquiring indoor point cloud data of a building through the laser scanner (1), preprocessing the point cloud data to remove interference points and generating a global point cloud model; the point cloud axial network generation module is used for identifying column point clouds in the global point cloud model based on an AI algorithm and extracting center points to generate an actual construction axial network; The point cloud and BIM model registration module is used for executing an axis network pair Ji Peizhun based on an actual construction axis network and a design axis network of a preset BIM model; The BIM component bounding box generation module is used for extracting three-dimensional parameters of the component to be monitored from the registered BIM model and generating BIM bounding box parameters based on the three-dimensional parameters; The construction state judging module is used for calculating an overlapping area of the global point cloud model and BIM bounding box parameters to obtain coverage data, calling a database which is pre-stored with a corresponding relation between a coverage threshold value and a construction state of the component, and searching the database based on the coverage data to obtain a matched construction state of the component; A construction progress plan importing module for importing a planned engineering amount and a planned construction period into the BIM model; the construction progress evaluation display module is used for counting the duty ratio of each state component according to the project, combining the planned project quantity and the planned construction period in the pre-led BIM model, calculating the progress deviation rate of the actual progress and the planned progress, and generating a visual evaluation report.

Description

Automatic assessment method and system for construction state of building indoor member Technical Field The application relates to the technical field of building construction, in particular to an automatic evaluation method and system for construction states of building indoor components. Background The construction engineering has the characteristics of long period, multiple links and strong uncertainty, and the construction progress management is always the key content of engineering management, and is an important link for ensuring the on-time completion of projects and the quality reaching standards. At present, in the progress evaluation of indoor construction of a building, a plurality of technical pain points exist in a mode of relying on manual inspection and manual recording for a long time, and the method is characterized in that the traditional manual measurement needs to check the size and the completion state of each component one by one, so that the data acquisition efficiency is low, the single progress evaluation needs to take 3 to 5 days for a large building comprising a structural component and an electromechanical pipeline, the real-time requirement of construction management is difficult to meet, the manual recording is easily influenced by subjective experience, and particularly, the condition that unified quantization standard is lacking when judging whether the component is completed or not is judged, the evaluation result deviation of different engineers in the same project can reach more than 15%, and the deviation is more prominent for the detailed components such as the electromechanical pipeline and the like, so that the evaluation precision is difficult to guarantee, therefore, the application provides a new technical scheme. Disclosure of Invention In order to realize the high efficiency and standardization of the construction progress evaluation of the building indoor components, the application provides an automatic construction state evaluation method and system for the building indoor components. In a first aspect, the application provides an automatic assessment method for construction states of indoor components of a building, which adopts the following technical scheme: An automatic assessment method for construction state of an indoor building component comprises the following steps: acquiring indoor point cloud data of a building, preprocessing the point cloud data to remove interference points and generating a global point cloud model; Identifying column point clouds in the global point cloud model based on an AI algorithm, extracting center points, and generating an actual construction shaft network; executing a shaft network pair Ji Peizhun based on the actual construction shaft network and a design shaft network of a preset BIM model; extracting three-dimensional parameters of a member to be monitored from the registered BIM model, and generating BIM bounding box parameters based on the three-dimensional parameters; calculating an overlapping region of the global point cloud model and BIM bounding box parameters to obtain coverage data, calling a database which is pre-stored with a corresponding relation between a coverage threshold value and a component construction state, and searching the database based on the coverage data to obtain a matched component construction state; And counting the duty ratio of each state component according to the itemized engineering, combining the planned engineering quantity and the planned construction period in the pre-led BIM model, calculating the progress deviation rate of the actual progress and the planned progress, and generating a visual evaluation report. Optionally, the preprocessing comprises the steps of removing interference points through a statistical filtering algorithm, preserving geometric features through voxel grid downsampling, and generating a globally consistent indoor point cloud model through a multi-station stitching algorithm based on overlapping point clouds. Optionally, the AI algorithm comprises automatically identifying a pillar point cloud region from a global point cloud model based on a pillar point cloud identification algorithm of PointCNN improved models, performing cylinder fitting on the identified pillar point cloud, and calculating a cylinder center coordinate as a pillar center point. Optionally, the generation of the actual construction shaft network comprises the steps of connecting adjacent column center points in pairs to generate longitudinal and transverse axes, selecting column center points around the building with the farthest span, connecting the column center points to form boundary axes, and constructing the complete actual construction shaft network. Optionally, the shaft network alignment registration comprises the steps of calculating an optimal transformation matrix of an actual construction shaft network and a design shaft network of the BIM model through a least square me