CN120633942A - A civil engineering cost management method based on artificial intelligence - Google Patents

A civil engineering cost management method based on artificial intelligence

Info

Publication number
CN120633942A
CN120633942A CN202510972020.3A CN202510972020A CN120633942A CN 120633942 A CN120633942 A CN 120633942A CN 202510972020 A CN202510972020 A CN 202510972020A CN 120633942 A CN120633942 A CN 120633942A
Authority
CN
China
Prior art keywords
volume
component
point cloud
project
actual
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.)
Granted
Application number
CN202510972020.3A
Other languages
Chinese (zh)
Other versions
CN120633942B (en
Inventor
代梦麟
洪城城
靳振军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Chong Nan Engineering Management Co ltd
Original Assignee
Hefei Yinyue Space Architectural Design Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hefei Yinyue Space Architectural Design Co ltd filed Critical Hefei Yinyue Space Architectural Design Co ltd
Priority to CN202510972020.3A priority Critical patent/CN120633942B/en
Publication of CN120633942A publication Critical patent/CN120633942A/en
Application granted granted Critical
Publication of CN120633942B publication Critical patent/CN120633942B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/24765—Rule-based classification
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/251—Fusion techniques of input or preprocessed data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/253—Fusion techniques of extracted features
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/08—Construction
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
    • G06V10/449—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters
    • G06V10/451—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters with interaction between the filter responses, e.g. cortical complex cells
    • G06V10/454—Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/70—Labelling scene content, e.g. deriving syntactic or semantic representations

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Business, Economics & Management (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Multimedia (AREA)
  • Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Geometry (AREA)
  • Molecular Biology (AREA)
  • Marketing (AREA)
  • Computational Linguistics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Biomedical Technology (AREA)
  • Computer Hardware Design (AREA)
  • Biophysics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Architecture (AREA)
  • Civil Engineering (AREA)
  • Primary Health Care (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Structural Engineering (AREA)
  • Computing Systems (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)

Abstract

本申请公开了一种基于人工智能的土木工程造价管理方法，涉及土木工程造价：构建施工项目的BIM设计模型；采用激光雷达采集施工现场的点云数据，生成实际施工点云模型；将实际施工点云模型与BIM设计模型进行空间配准，得到配准后的融合模型；根据融合模型，通过深度学习算法识别已完成构件，并通过体积计算，得到实际已完成工程量；比对实际已完成工程量与BIM设计模型中对应时间节点的计划完成工程量，计算进度偏差率；根据进度偏差率，结合工程量清单中的单价信息，调整工程造价。针对土木工程造价动态监控精度低，本申请通过激光点云与BIM模型融合，利用深度学习算法识别构件完成状态并计算实际工程量等，提高了工程造价的动态监控精度。

This application discloses a civil engineering cost management method based on artificial intelligence, which involves the following steps: constructing a BIM design model of a construction project; using a laser radar to collect point cloud data from the construction site to generate an actual construction point cloud model; spatially registering the actual construction point cloud model with the BIM design model to obtain a registered fusion model; based on the fusion model, identifying completed components through a deep learning algorithm, and obtaining the actual completed project volume through volume calculation; comparing the actual completed project volume with the planned completed project volume at the corresponding time node in the BIM design model to calculate the progress deviation rate; adjusting the project cost based on the progress deviation rate and the unit price information in the bill of quantities. In response to the low accuracy of dynamic monitoring of civil engineering cost, this application improves the dynamic monitoring accuracy of project cost by fusing laser point cloud with BIM model, using a deep learning algorithm to identify component completion status and calculate the actual project volume.

Description

Civil engineering cost management method based on artificial intelligence
Technical Field
The application relates to the field of engineering cost, in particular to a civil engineering cost management method based on artificial intelligence.
Background
The civil engineering project has the characteristics of large investment scale, long construction period and numerous participants, and engineering cost management runs through the full life cycle of the project, thereby being a key link for ensuring the economic benefit of the project. With the transition from the construction industry to digital and intelligent, the traditional cost management mode has difficulty in meeting the requirements of fine management of modern engineering projects. In particular, in the construction stage, the engineering completion condition is accurately mastered in real time, and the manufacturing cost plan is timely adjusted, so that the method has important significance in controlling project cost and avoiding investment risks. The application of the Building Information Model (BIM) technology provides a new technical means for engineering cost management, and realizes automatic extraction of engineering quantity and accurate calculation of the cost by constructing a three-dimensional digital model containing geometric information, material properties and cost data. However, the BIM model reflects design intent and plan state, and how to compare and analyze the actual construction state with the BIM model to realize the cost dynamic management based on the actual completion situation is still a technical problem facing the industry.
The traditional method relies on manual work to use tools such as a tape measure, a level gauge and the like to carry out field measurement, so that time and labor are consumed, and the measurement accuracy is greatly influenced by human factors. For parts which are difficult to reach, such as high-altitude operation, complex structures and the like, potential safety hazards exist in measurement work. Manual measurements are usually sampled and are difficult to cover all components, so that engineering quantity statistics are omitted. The existing method mostly adopts an estimation mode to determine the completion engineering quantity, such as roughly estimating the completion percentage according to the floor completion condition or reversely pushing the completion quantity according to the material consumption. This approach ignores the actual finished form of the component, and in particular, for partially finished components, does not allow accurate calculation of the finished volume. The reinforced concrete member is usually fixed by adopting fixed coefficients such as steel bar buckling and reducing, masonry mortar joint calculation and the like, and has deviation from the actual situation.
The collection and aggregation of engineering progress data is typically done monthly with significant hysteresis. When a progress deviation is found, the best adjustment opportunity is often missed. The relevance analysis among the sub projects is insufficient, and the progress risk affecting the critical path is difficult to discover in time. The progress information is disjointed with the cost data, so that the cost dynamic prediction based on the actual progress cannot be realized.
Disclosure of Invention
Aiming at low dynamic monitoring precision of civil engineering cost, the application provides an artificial intelligence-based civil engineering cost management method, which utilizes a deep learning algorithm to identify the completion state of a component and calculate the actual engineering quantity and the like through fusion of laser point cloud and a BIM model, thereby improving the dynamic monitoring precision of engineering cost.
The application provides a civil engineering cost management method based on artificial intelligence, which comprises the steps of S1, constructing a BIM design model of a construction project, wherein the BIM design model comprises component information, an engineering quantity list and planning progress data in a construction stage, S2, collecting point cloud data of a construction site by using a laser radar to generate an actual construction point cloud model, S3, carrying out spatial registration on the actual construction point cloud model and the BIM design model to obtain a registered fusion model, S4, identifying finished components according to the fusion model through a deep learning algorithm, and obtaining actual finished engineering quantity through volume calculation, S5, comparing the actual finished engineering quantity with the planning finished engineering quantity of a corresponding time node in the BIM design model, calculating a progress deviation rate, and S6, according to the progress deviation rate, combining unit price information in the engineering quantity list, and adjusting engineering cost.
Further, S4, identifying finished components through a deep learning algorithm according to a fusion model, and obtaining actual finished engineering quantity through volume calculation, wherein the method comprises the steps of extracting point cloud characteristic data from the fusion model, wherein the point cloud characteristic data comprise space coordinates, reflection intensity and normal vectors, carrying out voxelization processing on the point cloud characteristic data according to the space coordinates, converting continuous point cloud data into three-dimensional voxel grids, wherein each voxel unit comprises reflection intensity and normal vector information corresponding to a space position, inputting the three-dimensional voxel grids into a pre-trained convolutional neural network model to obtain component type labels, space positions and component point cloud clusters, matching the identified component point cloud clusters with standard components of corresponding positions and types in a BIM design model according to the component type labels and the space position information, and determining the finishing state of the components through calculating the point cloud coverage rate, wherein the finishing state comprises complete finishing, partial finishing and non-starting.
When the coverage rate of the point cloud is larger than a preset threshold value Q 1, the component is judged to be completely completed, the system obtains a unique identifier of the component in the BIM through a component matching result, directly inquires the designed volume attribute of the component through a data interface of the BIM, directly assigns the inquired designed volume value to be the actual completed volume, and the processing mode avoids calculation errors of the point cloud reconstruction and improves efficiency and accuracy.
When the point cloud coverage rate is smaller than a preset threshold value Q 1 and larger than a preset threshold value Q 2, determining that the component is partially completed, reconstructing the three-dimensional surface of the point cloud cluster of the component according to the space coordinates of the point cloud cluster of the corresponding component, calculating the volume contained in the closed surface, taking the calculated volume value as the actual completed volume of the component, and accumulating according to the engineering subsection according to the actual completed volume of each component and the volume calculation rule of the corresponding component type in the engineering quantity list to obtain the actual completed engineering quantity of each subsection engineering.
In particular, for the complete component R > Q 1, the application avoids the accumulated error in the process of processing the point cloud data by directly calling the accurate design volume in the BIM model. When the component is basically finished, the deviation between the actual geometry and the design geometry is within the engineering allowable range, and the design value is directly used, so that the precision is ensured, and the calculation efficiency is greatly improved. The component Q 2≤R≤Q1 is partially completed, the volume of the actually completed part is accurately calculated through a point cloud reconstruction technology, and the technical problem that the component in construction cannot be accurately estimated by the traditional method is solved. The component R < Q 2 is not started, so that the scattered noise points are prevented from being misjudged as the construction start, and the robustness of the system is improved.
Further, the spatial coordinates are used to determine the spatial location and geometry of the component, the reflected intensity is used to distinguish between components of different materials, and the normal vector is used to identify the surface orientation and boundaries of the component.
Further, the three-dimensional voxel grid is input into a pre-trained convolutional neural network model to obtain component type labels, space positions and component point cloud clusters, the method comprises the steps that the convolutional neural network model comprises a feature extraction layer, an instance segmentation layer and a semantic recognition layer, the feature extraction layer extracts geometric features and material features of components by utilizing reflection intensity and normal vector information, the instance segmentation layer separates adjacent components through a region growing algorithm based on the extracted geometric features and outputs the point cloud clusters of each independent component, the semantic recognition layer outputs the component type labels corresponding to the point cloud clusters based on the extracted material features and the geometric features, and the central position of each component is calculated according to boundary voxel coordinates of the component point cloud clusters to serve as the space positions.
Further, according to the component type tag and the spatial position information, matching each identified component point cloud cluster with a standard component of a corresponding position and type in the BIM design model, and determining the completion state of the component by calculating the point cloud coverage rate, wherein the method comprises the following steps:
Selecting the candidate component set of the same type in the BIM design model based on the component type label, calculating the barycenter coordinates of the component point cloud cluster as the center position C, Wherein n is the number of points in the component point cloud cluster, (x i,yi,zi) is the three-dimensional coordinate of the ith point, the distance D between the center position C and the geometric center of each component in the candidate component set is calculated,Selecting a component with the minimum distance D as a matching component, acquiring an axis alignment bounding box AABB parameter of the matching component, wherein the AABB parameter comprises a minimum vertex coordinate (x min,ymin,zmin) and a maximum vertex coordinate (x max,ymax,zmax), comparing each point coordinate in a component point cloud cluster with an axis alignment bounding box AABB boundary, counting the points N in falling into the axis alignment bounding box AABB, dividing the axis alignment bounding box AABB into MxNxK voxel units according to a preset resolution r, counting the points in each voxel unit, and calculating the actual distribution density
Calculating theoretical point cloud density rho standard by simulating a laser radar scanning process according to geometric information of a matching component in a BIM design model, calculating theoretical sampling points in unit volume according to scanning parameters of the laser radar and the surface area of the component, and calculating point cloud coverage rateWhen the apparent non-uniformity of the actual distribution density is detected, the point cloud is divided into a plurality of connected areas through a connected area analysis algorithm, the proportion epsilon of the points in the maximum connected area to N in is calculated, the coverage rate is corrected, wherein R' =R multiplied by epsilon, the coverage rate is judged to be completely finished when R > Q 1, the coverage rate is judged to be partially finished when Q 2≤R≤Q1, the coverage rate is judged to be not started when R < Q 2, and Q 1 and Q 2 are preset coverage rate thresholds. The value range of Q 1 is 90-100%, and the value range of Q 2 is 10-20%.
In particular, first, based on the deeply learning identified component type tags, the matching search space is reduced from thousands of components of the full model to tens of components of the same type, and the search complexity is reduced from O (n) to O (n/k), where k is the number of component types. The introduction of the semantic priori knowledge avoids 'alien and alien' mismatching which can occur in pure geometric matching.
Second, centroidAs the geometric feature center of the point cloud cluster, there is translational invariance and noise robustness. By Euclidean distanceConducting the nearest neighbor search in principle ensures that the components closest in spatial position are correctly matched. The centroid-based matching method is high in calculation efficiency, the complexity is only O (m), and m is the number of candidate components.
And finally, calculating rho standard by simulating a laser radar scanning process, and considering the influence of actual factors such as scanning angle, distance, shielding and the like on the point cloud density. The theoretical value calculation based on the physical model is closer to the actual condition than the simple uniform distribution assumption, and the coverage rate consistency problem under different scanning conditions is solved.
Further, when the point cloud coverage rate is smaller than a preset threshold value Q 1 and larger than a preset threshold value Q 2, judging that the component is partially completed, reconstructing the three-dimensional surface of the point cloud cluster of the component according to the space coordinates of the point cloud cluster of the corresponding component, calculating the volume contained in the closed surface, taking the calculated volume value as the actual completed volume of the component, wherein the method comprises the steps of performing outlier processing on the point cloud cluster of the partially completed component to obtain effective point cloud, extracting construction direction information of the matched component from a BIM design model to serve as a main construction direction vector, wherein civil engineering construction has definite directional characteristics, namely column components are vertically constructed from bottom to top, beam components are horizontally constructed along the long axis direction, and plate components are layered and poured from bottom to top. The method comprises the steps of reading type attributes and geometric parameters of components from a BIM model, automatically determining a main construction direction vector according to the types of the components, and reading a self-defined direction from construction information attributes of the BIM model for special components.
And carrying out one-dimensional projection on the three-dimensional point cloud data along the main construction direction vector, and simplifying the space distribution problem into a linear distribution problem. The projected value of each point in the construction direction represents its position in the construction process. The method comprises the steps of determining a range interval of projection values, namely a distance from a construction starting point to a current farthest point, equally dividing the interval into m subintervals (the m value is adaptively adjusted according to the size of a component and is generally 30-50), counting the number of points in each subinterval, dividing the length of the interval to obtain a linear density value, and forming a density distribution curve along the construction direction.
In addition, at the construction interface position, the completed part of the point cloud is dense, the incomplete part of the point cloud is sparse or no point cloud exists, and therefore the characteristic of sharp reduction of the density occurs at the interface. Calculating the density ratio of adjacent subintervals one by one from the construction starting end, and when the density ratio of a certain position is smaller than a threshold value T (empirical value 0.3), indicating that the density is suddenly changed, wherein the suddenly changed position is the construction completion interface.
The cutting plane passes through the identified construction interface position points, the plane normal vector is the main construction direction vector, and the cutting plane is ensured to divide the component into two parts which are completed and unfinished. And calling a built-in geometric engine (such as ACIS, parasolid and the like) of BIM software, executing intersection operation of the entity and the half space, and reserving the geometric body at one side of the construction starting end. The cut geometry is a standard B-Rep solid model, the volume calculation function of the geometry engine is directly called, the obtained volume value is an accurate mathematical calculation result, no accumulated error exists, and the volume value is used as the actual finished volume of the partially finished component.
In particular, the application converts the conventional point cloud surface reconstruction problem into a geometric cutting problem based on a BIM model. Traditional point cloud surface reconstruction, such as the alpha shape algorithm, needs to infer the complete geometry from discrete point clouds, belongs to the reverse engineering from part to whole, and inevitably has information deletion and reconstruction errors. The method directly uses the accurate geometry provided by the BIM model, simplifies the problem into 'determining the cutting position', belongs to 'forward analysis from whole to part', and avoids uncertainty of geometric reconstruction in principle. In addition, alpha parameter selection of the alpha shape algorithm is subjective, different alpha values can generate different boundary shapes, and the reconstructed boundary cannot be ensured to be consistent with an actual construction interface. According to the scheme, the construction interface is directly positioned through density analysis, the standard cutting plane is built, the three-dimensional boundary recognition problem is reduced to be the one-dimensional interface detection problem, and the uniqueness and the accuracy of the boundary are ensured in principle.
Further, according to the volume calculation rule of the actual completion volume of each component and the corresponding component type in the engineering quantity list, the actual completed engineering quantity of each component is obtained by accumulating engineering component items, wherein the actual completion engineering quantity is calculated according to the following rule of beam and column components, namely actual completion engineering quantity = actual completion volume x (1-0.025), wherein 0.025 is a steel bar volume deduction coefficient, plate components, namely actual completion engineering quantity = actual completion volume x (1-0.015), wherein 0.015 is a steel bar volume deduction coefficient, the actual completion engineering quantity = actual completion volume, and the steel bar volume is not deducted, and the actual completion engineering quantity = actual completion volume x 7850kg/m 3/1000 is calculated according to the following rule of steel structure components.
Further, the volume calculation rule further comprises the steps of calculating a standard brick masonry according to the following rule, namely, actual completion engineering quantity = actual completion volume multiplied by 0.95, wherein 0.95 is a reduction coefficient considering mortar joints, the brick masonry comprises actual completion engineering quantity = actual completion volume multiplied by 1.1, the number of bricks is obtained, 1.1 is a mortar joint coefficient, and the template engineering is calculated according to the following rule, namely, beam template comprises actual completion engineering quantity= (beam bottom width +2 x beam height) multiplied by actual completion length, plate template comprises actual completion engineering quantity = actual completion area, and all component engineering quantities under the same sub engineering code are accumulated according to the same metering unit, so that the actual completion engineering quantity of the sub engineering is obtained.
Further, S5, comparing the actual completed engineering quantity with the planned completed engineering quantity of the corresponding time node in the BIM design model, and calculating a progress deviation rate, wherein the progress deviation rate comprises the steps of obtaining the current construction date, extracting the planned completed engineering quantity of the corresponding time node from the planned progress data of the BIM design model, comparing the actual completed engineering quantity of each project with the corresponding planned completed engineering quantity, calculating the progress deviation rate of each project, namely a progress deviation rate= (actual completed engineering quantity-planned completed engineering quantity)/planned completed engineering quantity multiplied by 100%, calculating a weighted average progress deviation rate according to the weight coefficient of each project in the total project, wherein the weight coefficient is determined according to the contract amount occupation ratio of each project, generating progress early warning information when the progress deviation rate of a project exceeds a preset threshold value, wherein the early warning information comprises a project name, a deviation rate numerical value and a deviation reason analysis, judging the progress state according to the positive and negative values of the progress deviation rate, representing the progress deviation rate when the progress deviation rate is positive, and generating a report form, and accumulating the progress deviation rate according to the difference value, namely the progress deviation rate and the actual report form and the progress deviation.
Compared with the prior art, the application has the advantages that:
According to the application, the high-density point cloud data (resolution reaches millimeter level) acquired by the laser radar and the BIM design model are accurately registered and fused, and the three-dimensional convolutional neural network is utilized to automatically identify the components and calculate the point cloud coverage rate, so that the accurate calculation of the engineering quantity is realized.
For the complete component (coverage rate R > 90%), directly extracting the accurate design volume in the BIM model, avoiding the point cloud reconstruction error, and for the partial complete component (R is more than or equal to 10% and less than or equal to 90%), accurately identifying the construction interface through density gradient analysis, and obtaining the accurate volume by using the BIM geometric engine to carry out Boolean cutting operation.
Compared with the traditional manual measurement and estimation method, the method reduces the engineering calculation error from 10% -15% to 2% -3%, and particularly realizes full-coverage and high-precision automatic measurement for complex special-shaped components and high-altitude parts which are difficult to reach. Meanwhile, a fine engineering quantity calculation rule is established for different material types, such as differential buckling coefficient (beam column 2.5%, plate 1.5%) of reinforced concrete members, and accurate conversion of masonry mortar joints, so that accurate conversion from actual finished volume to engineering quantity is ensured, and a reliable data base is provided for dynamic monitoring of manufacturing cost.
Drawings
The application will be further described by way of exemplary embodiments, which will be described in detail by way of the accompanying drawings. The embodiments are not limiting, in which like numerals represent like structures, wherein:
FIG. 1 is an exemplary flow chart of a method of civil engineering cost management based on artificial intelligence, according to some embodiments of the application;
FIG. 2 is an exemplary flow chart for obtaining an actual amount of work done according to some embodiments of the application;
FIG. 3 is an exemplary flow chart for generating component point cloud clusters according to some embodiments of the application;
FIG. 4 is an exemplary flow chart of computing point cloud coverage according to some embodiments of the application.
Detailed Description
The method and system provided by the embodiment of the application are described in detail below with reference to the accompanying drawings.
The method comprises the steps of 1, constructing a BIM design model of a construction project, wherein the BIM design model comprises component information, an engineering quantity list and planning progress data in a construction stage, 2, collecting point cloud data of a construction site by using a laser radar to generate an actual construction point cloud model, 3, carrying out spatial registration on the actual construction point cloud model and the BIM design model to obtain a registered fusion model, 4, identifying finished components according to the fusion model through a deep learning algorithm and obtaining an actual finished engineering quantity through volume calculation, 5, comparing the actual finished engineering quantity with the planning finished engineering quantity of a corresponding time node in the BIM design model, calculating a progress deviation rate, and 6, according to the progress deviation rate, combining the unit price information in the engineering quantity list, and adjusting engineering cost.
Specifically, S1, a BIM design model of the construction project is constructed, where the BIM design model includes component information, an engineering quantity list, and planning progress data of the construction stage.
S2, acquiring point cloud data of a construction site by using a laser radar to generate an actual construction point cloud model, performing 360-degree omnibearing scanning on the construction site by using a laser radar scanner according to a preset time interval, preprocessing the acquired original point cloud data, including denoising, registering and dividing, classifying the point cloud data by using a semantic dividing algorithm based on deep learning, identifying different materials and component types, and generating the actual construction point cloud model with semantic tags.
S3, performing spatial registration on an actual construction point cloud model and a BIM design model to obtain a registered fusion model, wherein the spatial registration comprises the steps of extracting characteristic geometric elements of a building from the BIM design model, including wall corner points, column center points and floor boundary lines, generating a BIM characteristic point set, detecting plane characteristics from the actual construction point cloud model through a RANSAC algorithm, identifying wall surfaces, building boards and columns, extracting plane intersection lines and intersection points as a point cloud characteristic point set, performing preliminary matching on the BIM characteristic point set and the point cloud characteristic point set, calculating an initial transformation matrix T 0 by adopting a closest point iterative algorithm, including a rotation matrix R 0 and a translation vector T 0, performing coarse registration on the actual construction point cloud model based on the initial transformation matrix T 0, converting the point cloud coordinate system into the BIM coordinate system, performing fine registration on the basis of the coarse registration by adopting a point-to-face ICP algorithm, namely projecting each point in the point cloud to the nearest component surface in the BIM model, calculating the point-to-face distance, and constructing an error functionThe method comprises the steps of obtaining a point cloud model of the actual construction, wherein d i is the distance from an ith point to a corresponding surface, iteratively optimizing transformation parameters through a least square method until an error function converges, transforming the actual construction point cloud model by applying a final transformation matrix T final to obtain a registration point cloud under the same coordinate system with a BIM design model, integrating registered point cloud data with geometric data of the BIM design model, and generating a fusion model containing design information and actual construction information.
As shown in FIG. 2, S4, according to the fusion model, the completed component is identified through a deep learning algorithm, and the actual completed engineering quantity is obtained through volume calculation, point cloud characteristic data are extracted from the fusion model, the point cloud characteristic data comprise space coordinates, reflection intensity and normal vectors, the space coordinates are used for determining the space position and the geometric shape of the component, the reflection intensity is used for distinguishing the components of different materials, and the normal vectors are used for identifying the surface orientation and the boundary of the component. Wherein the spatial coordinates of each point P i are represented as (x i,yi,zi), where i [ E ] [1, N ], N is the total number of point clouds, and the reflection intensityWherein, P r is the receiving power, P e is the transmitting power, the reflection intensity characteristic values of different materials are concrete I i E [0.3,0.5], steel I i E [0.7,0.9], brick masonry I i E [0.4,0.6], and the reflection intensity is clustered by Gaussian Mixture Model (GMM) to realize automatic identification of materials.
Calculating a vector, calculating a normal vector N i＝(nx,ny,nz of each point by adopting a Principal Component Analysis (PCA) algorithm, selecting a k neighbor point set N k(Pi) of a point P i, and constructing a covariance matrix by using k=20Wherein p j∈Nk(Pi), normal vector n i is the eigenvector corresponding to the minimum eigenvalue of C. When the normal vector included angle theta of the adjacent points is more than 30 degrees, the boundary is judged.
And (3) carrying out voxelization processing on the point cloud characteristic data according to the space coordinates, converting continuous point cloud data into a three-dimensional voxel grid, wherein each voxel unit comprises reflection intensity and normal vector information corresponding to the space position, specifically, setting voxel resolution v r =0.05 m (5 cm), and calculating a point cloud bounding box: [ x min,xmax]×[ymin,ymax]×[zmin,zmax ]. Generating voxel grid dimensions:
The three-dimensional voxel grid is input into a pre-trained convolutional neural network model to obtain component type labels, space positions and component point cloud clusters, the convolutional neural network model comprises a feature extraction layer, an example segmentation layer and a semantic recognition layer, the feature extraction layer extracts geometric features and material features of components by utilizing reflection intensity and normal vector information, the example segmentation layer separates adjacent components through a region growing algorithm based on the extracted geometric features and outputs the point cloud clusters of each independent component, the semantic recognition layer outputs the component type labels corresponding to each point cloud cluster based on the extracted material features and the geometric features, and the central position of each component is calculated according to boundary voxel coordinates of the component point cloud clusters to serve as the space positions.
As shown in fig. 3 and 4, each identified component point cloud cluster is matched with a standard component of a corresponding position and type in the BIM design model according to component type tags and spatial position information, the completion status of the component is determined by calculating the point cloud coverage rate, the completion status includes complete completion, partial completion and non-start, candidate component sets of the same type are screened in the BIM design model based on the component type tags, centroid coordinates of the component point cloud cluster are calculated as a center position C,Wherein n is the number of points in the component point cloud cluster, (x i,yi,zi) is the three-dimensional coordinate of the ith point, the distance D between the center position C and the geometric center of each component in the candidate component set is calculated,
Selecting a component with the minimum distance D as a matching component, acquiring an axis alignment bounding box AABB parameter of the matching component, wherein the AABB parameter comprises a minimum vertex coordinate (x min,ymin,zmin) and a maximum vertex coordinate (x max,ymax,zmax), comparing each point coordinate in a component point cloud cluster with an axis alignment bounding box AABB boundary, counting the points N in falling into the axis alignment bounding box AABB, dividing the axis alignment bounding box AABB into MxNxK voxel units according to a preset resolution r, counting the points in each voxel unit, and calculating the actual distribution densityCalculating theoretical point cloud density rho standard by simulating a laser radar scanning process according to geometric information of a matching component in a BIM design model, calculating theoretical sampling points in unit volume according to scanning parameters of the laser radar and the surface area of the component, and calculating point cloud coverage rateWhen the apparent non-uniformity of the actual distribution density is detected, the point cloud is divided into a plurality of connected areas through a connected area analysis algorithm, the proportion epsilon of the points in the largest connected area to Nin is calculated, the coverage rate is corrected, wherein R' =R multiplied by epsilon, the complete completion is judged when R > Q 1, the partial completion is judged when Q 2≤R≤Q1, the non-start is judged when R < Q 2, and Q 1 and Q 2 are preset coverage rate thresholds. The value range of Q 1 is 90-100%, and the value range of Q 2 is 10-20%.
When the coverage rate R > Q 1(Q1 =0.95), the component is judged to be completely completed, the unique mark of the component is CID (ComponentID), the design volume is extracted through a BIM data interface function, wherein V actual =BIM.GetVolume (CID), V actual represents the actual completed volume, the unit is m 3, and the design volume is directly used as the actual completed volume, so that the point cloud reconstruction error is avoided.
When Q 2≤R≤Q1(Q2＝0.3,Q1 =0.95), a refinement process flow is performed.
Outliers are first culled and for each point P i in the point cloud cluster p= { P 1,p2,.....,pn }, the average distance of P i to its k nearest neighbors is calculated: where k=20 (empirical value), calculating the statistical features of the average distance of all points: outlier determination criterion, if Then P i is the outlier, and all outliers are removed, resulting in an effective point cloud P valid.
Extracting a main construction direction vector, and extracting a main construction direction vector v from the BIM model according to the component type T, wherein if T= "column": v= (0, 1) T, and if T= "beam": If t= "plate": v= (0, 1) T, if t= "wall": v=n wall (wall normal vector), where p start,pend is the start and end coordinates of the beam, respectively.
Projecting each point P i＝(xi,yi,zi)T in the effective point cloud P valid to t i＝pi×v＝xivx+yivy+zivz, and determining a projection interval ：[tmin,tmax];tmin＝min{ti|i＝1,2,.....,n};tmax＝max{ti|i＝1,2,.....,n}.
The projection interval is equally divided into m subintervals (m=50): the j-th subinterval: I j＝[tmin+(j-1)Δt,tmin + j deltat ], j=1, 2, once again, m. Counting the number of points of each subinterval, namely n j＝|{pi|ti∈Ij, and calculating the linear density: (Unit: points/meter).
Identifying the construction interface position, and calculating the density ratio of adjacent subintervals: The construction interface determines that if r j < T (t=0.3) and ρ j+1＜ρmin, the interface is at the end of the j-th subinterval, where ρ min＝0.1×max{ρj |j=1, 2.
The partially completed volume was cut and calculated, cut point: P cut＝porigin+hcut ×v, where P origin is the component starting point. Cutting plane equation v× (P-P cut) =0, expanded form v x(x-xcut)+vy(y-ycut)+vz(z-zcut) =0.
Boolean cut operation, defining half space H.H= { P|v× (P-P cut). Ltoreq.0 }. Performing intersection operations:
G partial＝GBIM n H, wherein G BIM is the building block geometry in BIM and G partial is the cut portion. The Volume calculation is V actual＝∫∫∫Gpartial dV, and the accurate value is obtained through the Volume () function of the geometric engine.
And according to the actual finished volume of each component and the volume calculation rule of the corresponding component type in the engineering quantity list, accumulating according to the engineering subsection items to obtain the actual finished engineering quantity of each subsection engineering.
For the reinforced concrete member, the beam and column members are calculated according to the following rule that the actual completion engineering quantity=the actual completion volume (1-0.025), wherein 0.025 is the volume deduction coefficient of the reinforced steel bars, the plate members are calculated according to the following rule that the actual completion engineering quantity=the actual completion volume (1-0.015), wherein 0.015 is the volume deduction coefficient of the reinforced steel bars, and the base members are calculated according to the following rule that the actual completion engineering quantity=the actual completion volume, and the reinforced steel bar volume is not deducted.
For steel structural members, the actual amount of work done = actual volume done x 7850kg/m 3/1000 is calculated according to the following rule.
For the masonry member, a standard brickwork is calculated according to the following rule that the actual completion engineering amount = actual completion volume x 0.95, wherein 0.95 is a reduction coefficient considering mortar joints, and a block masonry is calculated according to the following rule that the actual completion engineering amount = actual completion volume ≡ (block volume x 1.1), the number of blocks is obtained, wherein 1.1 is a mortar joint coefficient.
For the template engineering, calculating according to the following rules, namely, a beam template, wherein the actual completion engineering quantity= (beam bottom width+2×beam height) ×the actual completion length, a plate template, namely, the actual completion engineering quantity = the actual completion area, accumulating all component engineering quantities under the same subentry engineering code according to the same metering unit to obtain the actual completion engineering quantity of the subentry;
S5, comparing the actual completed engineering quantity with the planned completed engineering quantity of the corresponding time node in the BIM design model, and calculating a progress deviation rate;
Comparing the actual completed engineering quantity of each sub-project with the corresponding planned completed engineering quantity, and calculating the progress deviation rate of each sub-project, wherein the progress deviation rate is = (actual completed engineering quantity-planned completed engineering quantity)/planned completed engineering quantity multiplied by 100%;
Calculating a weighted average progress deviation rate according to the weight coefficient of each sub-project in the total project, wherein the weight coefficient is determined according to the contract amount ratio of each sub-project;
And judging the progress state of the project according to the positive and negative values of the progress deviation rate, wherein the progress state represents the advance of the project when the progress deviation rate is positive, the progress hysteresis when the progress deviation rate is negative, and a progress deviation analysis report is generated, wherein the progress deviation analysis report comprises the planned project quantity, the actual project quantity, the deviation rate and the accumulated completion percentage of each sub project.
And S6, according to the progress deviation rate, combining the unit price information in the engineering quantity list, and adjusting the engineering cost.
The foregoing has been described schematically the application and embodiments thereof, which are not limiting, but are capable of other specific forms of implementing the application without departing from its spirit or essential characteristics. The drawings are merely illustrative of one embodiment of the present application, and the actual structure is not limited thereto. Therefore, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical scheme are not creatively designed without departing from the gist of the present application. In addition, the word "comprising" does not exclude other elements or steps, and the word "a" or "an" preceding an element does not exclude the inclusion of a plurality of such elements. The terms first, second, etc. are used to denote a name, but not any particular order.

Claims (10)

1.一种基于人工智能的土木工程造价管理方法，其特征在于，包括：1. A civil engineering cost management method based on artificial intelligence, characterized by comprising: S1，构建施工项目的BIM设计模型，BIM设计模型包括施工阶段的构件信息、工程量清单和计划进度数据；S1, build a BIM design model for the construction project, which includes component information, bill of quantities, and planned progress data for the construction phase; S2，采用激光雷达采集施工现场的点云数据，生成实际施工点云模型；S2, uses LiDAR to collect point cloud data of the construction site and generate a point cloud model of the actual construction; S3，将实际施工点云模型与BIM设计模型进行空间配准，得到配准后的融合模型；S3, spatially registering the actual construction point cloud model with the BIM design model to obtain a registered fusion model; S4，根据融合模型，通过深度学习算法识别已完成构件，并通过体积计算，得到实际已完成工程量；S4, based on the fusion model, identifies completed components through deep learning algorithms and calculates the actual completed project volume; S5，比对实际已完成工程量与BIM设计模型中对应时间节点的计划完成工程量，计算进度偏差率；S5, compare the actual completed project volume with the planned completed project volume at the corresponding time node in the BIM design model, and calculate the progress deviation rate; S6，根据进度偏差率，结合工程量清单中的单价信息，调整工程造价。S6, adjust the project cost based on the progress deviation rate and the unit price information in the bill of quantities. 2.根据权利要求1所述的基于人工智能的土木工程造价管理方法，其特征在于：2. The civil engineering cost management method based on artificial intelligence according to claim 1 is characterized in that: S4，根据融合模型，通过深度学习算法识别已完成构件，并通过体积计算，得到实际已完成工程量，包括：S4, based on the fusion model, uses a deep learning algorithm to identify completed components and calculates the actual completed project volume through volume calculation, including: 从融合模型中提取点云特征数据，点云特征数据包括空间坐标、反射强度和法向量；Extracting point cloud feature data from the fusion model, where the point cloud feature data includes spatial coordinates, reflection intensity, and normal vector; 根据空间坐标对点云特征数据进行体素化处理，将连续的点云数据转换为三维体素网格，每个体素单元包含对应空间位置的反射强度和法向量信息；The point cloud feature data is voxelized according to the spatial coordinates, and the continuous point cloud data is converted into a three-dimensional voxel grid. Each voxel unit contains the reflection intensity and normal vector information of the corresponding spatial position; 将三维体素网格输入预训练的卷积神经网络模型中，得到构件类型标签、空间位置和构件点云簇；The 3D voxel grid is input into a pre-trained convolutional neural network model to obtain component type labels, spatial locations, and component point cloud clusters. 根据构件类型标签和空间位置信息，将识别出的各个构件点云簇与BIM设计模型中对应位置和类型的标准构件进行匹配，通过计算点云覆盖率确定构件的完成状态，完成状态包括完全完成、部分完成和未开始；Based on the component type label and spatial location information, the identified component point cloud clusters are matched with the standard components of the corresponding position and type in the BIM design model. The completion status of the component is determined by calculating the point cloud coverage. The completion status includes fully completed, partially completed, and not started. 当点云覆盖率大于预设阈值Q1时，判定为完全完成构件，直接从BIM设计模型中提取对应构件的设计体积作为实际完成体积；When the point cloud coverage is greater than the preset threshold Q 1 , the component is determined to be completely completed, and the design volume of the corresponding component is directly extracted from the BIM design model as the actual completed volume; 当点云覆盖率小于预设阈值Q1且大于预设阈值Q2时，判定为部分完成构件，根据对应构件的点云簇空间坐标，重建构件点云簇的三维表面，计算封闭表面所包含的体积，将计算得到的体积值作为该构件的实际完成体积；When the point cloud coverage is less than the preset threshold Q1 and greater than the preset threshold Q2 , it is determined to be a partially completed component. According to the spatial coordinates of the point cloud cluster of the corresponding component, the three-dimensional surface of the component point cloud cluster is reconstructed, and the volume contained in the closed surface is calculated. The calculated volume value is used as the actual completed volume of the component; 根据各构件的实际完成体积和工程量清单中对应构件类型的体积计算规则，按照工程分部分项累加得到各分项工程的实际已完成工程量。According to the actual completed volume of each component and the volume calculation rules of the corresponding component type in the bill of quantities, the actual completed volume of each sub-project is obtained by adding up the sub-items of the project. 3.根据权利要求2所述的基于人工智能的土木工程造价管理方法，其特征在于：3. The civil engineering cost management method based on artificial intelligence according to claim 2 is characterized in that: 空间坐标用于确定构件的空间位置和几何形状；Spatial coordinates are used to determine the spatial position and geometric shape of components; 反射强度用于区分不同材质的构件；Reflection intensity is used to distinguish components of different materials; 法向量用于识别构件的表面朝向和边界。Normal vectors are used to identify the surface orientation and boundaries of components. 4.根据权利要求2所述的基于人工智能的土木工程造价管理方法，其特征在于：4. The civil engineering cost management method based on artificial intelligence according to claim 2 is characterized in that: 将三维体素网格输入预训练的卷积神经网络模型中，得到构件类型标签、空间位置和构件点云簇，包括：The 3D voxel grid is fed into a pre-trained convolutional neural network model to obtain component type labels, spatial locations, and component point cloud clusters, including: 卷积神经网络模型包括特征提取层、实例分割层和语义识别层；The convolutional neural network model includes a feature extraction layer, an instance segmentation layer, and a semantic recognition layer; 特征提取层利用反射强度和法向量信息提取构件的几何特征和材质特征；The feature extraction layer uses reflection intensity and normal vector information to extract the geometric features and material features of the component; 实例分割层基于提取的几何特征，通过区域生长算法分离相邻构件，输出各个独立构件的点云簇；The instance segmentation layer separates adjacent components through the region growing algorithm based on the extracted geometric features and outputs point cloud clusters of each independent component; 语义识别层基于提取的材质特征和几何特征，输出各点云簇对应的构件类型标签；The semantic recognition layer outputs the component type label corresponding to each point cloud cluster based on the extracted material features and geometric features; 根据构件点云簇的边界体素坐标，计算每个构件的中心位置，作为空间位置。According to the boundary voxel coordinates of the component point cloud cluster, the center position of each component is calculated as the spatial position. 5.根据权利要求2所述的基于人工智能的土木工程造价管理方法，其特征在于：5. The civil engineering cost management method based on artificial intelligence according to claim 2 is characterized in that: 根据构件类型标签和空间位置信息，将识别出的各个构件点云簇与BIM设计模型中对应位置和类型的标准构件进行匹配，通过计算点云覆盖率确定构件的完成状态，包括：Based on the component type label and spatial location information, the identified component point cloud clusters are matched with the standard components of the corresponding position and type in the BIM design model. The completion status of the component is determined by calculating the point cloud coverage, including: 基于构件类型标签，在BIM设计模型中筛选相同类型的候选构件集合；Based on component type labels, filter candidate component sets of the same type in the BIM design model; 计算构件点云簇的质心坐标，作为中心位置C，其中，n为构件点云簇中的点数，(xi,yi,zi)为第i个点的三维坐标；Calculate the centroid coordinates of the component point cloud cluster as the center position C, Where n is the number of points in the component point cloud cluster, (x i , y i , z i ) is the three-dimensional coordinate of the i-th point; 计算中心位置C与候选构件集合中各构件几何中心的距离D，Calculate the distance D between the center position C and the geometric center of each component in the candidate component set, 选择距离D最小的构件作为匹配构件； Select the component with the smallest distance D as the matching component; 获取匹配构件的轴对齐包围盒AABB参数，AABB参数包括最小顶点坐标(xmin,ymin,zmin)和最大顶点坐标(xmax,ymax,zmax)；Get the axis-aligned bounding box (AABB) parameters of the matching component. The AABB parameters include the minimum vertex coordinates (x min , y min , z min ) and the maximum vertex coordinates (x max , y max , z max ); 将构件点云簇中的每个点坐标与轴对齐包围盒AABB边界进行比较，统计落入轴对齐包围盒AABB的点数Nin；Compare the coordinates of each point in the component point cloud cluster with the boundary of the axis-aligned bounding box AABB, and count the number of points Nin that fall into the axis-aligned bounding box AABB; 将轴对齐包围盒AABB按照预设分辨率r划分为M×N×K个体素单元，统计每个体素单元内的点数，计算实际分布密度 Divide the axis-aligned bounding box AABB into M×N×K voxel units according to the preset resolution r, count the number of points in each voxel unit, and calculate the actual distribution density 根据匹配构件在BIM设计模型中的几何信息，通过模拟激光雷达扫描过程计算理论点云密度ρstandard；根据激光雷达的扫描参数和构件表面积，计算单位体积内的理论采样点数；Based on the geometric information of the matching component in the BIM design model, the theoretical point cloud density ρ standard is calculated by simulating the laser radar scanning process. The theoretical number of sampling points per unit volume is calculated based on the laser radar scanning parameters and the surface area of the component. 计算点云覆盖率R＝ρactual ×100％，当检测到实际分布密度存在明显不均匀时，通过ρstandard Calculate the point cloud coverage R = ρ actual × 100%. When the actual distribution density is detected to be obviously uneven, the ρ standard 连通域分析算法将点云划分为多个连通区域，计算最大连通域内的点数占Nin的比例ε，对覆盖率进行修正：R'＝R×ε；The connected domain analysis algorithm divides the point cloud into multiple connected regions, calculates the ratio ε of the number of points in the largest connected domain to N in , and corrects the coverage: R' = R × ε; 当R＞Q1时判定为完全完成，当Q2≤R≤Q1时判定为部分完成，当R＜Q2时判定为未开始，其中，Q1和Q2为预设的覆盖率阈值。When R> Q1 , it is determined to be fully completed; when Q2≤R≤Q1 , it is determined to be partially completed; when R< Q2 , it is determined to be not started, where Q1 and Q2 are preset coverage thresholds. 6.根据权利要求5所述的基于人工智能的土木工程造价管理方法，其特征在于：6. The civil engineering cost management method based on artificial intelligence according to claim 5 is characterized by: Q1的取值范围90％至100％；The value range of Q 1 is 90% to 100%; Q2的取值范围10％至20％。The value of Q2 ranges from 10% to 20%. 7.根据权利要求4所述的基于人工智能的土木工程造价管理方法，其特征在于：7. The civil engineering cost management method based on artificial intelligence according to claim 4 is characterized in that: 当点云覆盖率小于预设阈值Q1且大于预设阈值Q2时，判定为部分完成构件，根据对应构件的点云簇空间坐标，重建构件点云簇的三维表面，计算封闭表面所包含的体积，将计算得到的体积值作为该构件的实际完成体积，包括：When the point cloud coverage is less than the preset threshold Q1 and greater than the preset threshold Q2 , the component is determined to be partially completed. Based on the spatial coordinates of the point cloud cluster of the corresponding component, the three-dimensional surface of the component point cloud cluster is reconstructed, and the volume contained in the closed surface is calculated. The calculated volume value is used as the actual completed volume of the component, including: 对部分完成构件的点云簇进行离群点处理，得到有效点云；Perform outlier processing on the point cloud cluster of the partially completed component to obtain a valid point cloud; 从BIM设计模型中提取匹配构件的施工方向信息，作为主施工方向向量；Extract the construction direction information of the matching components from the BIM design model as the main construction direction vector; 沿主施工方向向量对有效点云进行投影，将投影区间等分为m个子区间，统计每个子区间内的点数，计算有效点云在主施工方向上的密度分布；Project the valid point cloud along the main construction direction vector, divide the projection interval into m subintervals, count the number of points in each subinterval, and calculate the density distribution of the valid point cloud in the main construction direction; 计算相邻子区间的密度比值，当密度比值ρi 小于预设阈值T时，将第i个子区间的末端ρi+1 Calculate the density ratio of adjacent subintervals. When the density ratio ρ i is less than the preset threshold T, move the end ρ i+1 of the i-th subinterval to the 位置确定为施工界面位置；The location is determined as the construction interface location; 根据施工界面位置，在BIM设计模型中匹配构件的几何模型上设置切割平面，切割平面经过施工界面位置且垂直于主施工方向向量；According to the construction interface position, a cutting plane is set on the geometric model of the matching component in the BIM design model. The cutting plane passes through the construction interface position and is perpendicular to the main construction direction vector. 利用BIM中的几何引擎对匹配构件的几何模型进行布尔切割运算，保留施工起始端至切割平面之间的部分，计算保留部分的体积，作为部分完成构件的实际完成体积。The geometry engine in BIM is used to perform Boolean cutting operations on the geometric models of the matching components, retaining the part between the construction start end and the cutting plane, and calculating the volume of the retained part as the actual completed volume of the partially completed component. 8.根据权利要求7所述的基于人工智能的土木工程造价管理方法，其特征在于：8. The artificial intelligence-based civil engineering cost management method according to claim 7, characterized in that: 根据各构件的实际完成体积和工程量清单中对应构件类型的体积计算规则，按照工程分部分项累加得到各分项工程的实际已完成工程量，包括：Based on the actual completed volume of each component and the volume calculation rules of the corresponding component type in the bill of quantities, the actual completed volume of each sub-project is obtained by accumulating the sub-items of the project, including: 对于钢筋混凝土构件，按照如下规则计算：For reinforced concrete components, the calculation is based on the following rules: 梁、柱构件：实际完成工程量＝实际完成体积×(1-0.025)，其中，0.025为钢筋体积扣减系数；Beam and column components: Actual completed project quantity = actual completed volume × (1-0.025), where 0.025 is the steel bar volume deduction factor; 板类构件：实际完成工程量＝实际完成体积×(1-0.015)，其中，0.015为钢筋体积扣减系数；Plate components: Actual completed project volume = actual completed volume × (1-0.015), where 0.015 is the steel bar volume deduction factor; 基础构件：实际完成工程量＝实际完成体积，不扣除钢筋体积；Foundation components: actual completed project volume = actual completed volume, excluding steel bar volume; 对于钢结构构件，按照如下规则计算：For steel structural members, calculations are carried out according to the following rules: 实际完成工程量＝实际完成体积×7850kg/m3÷1000。Actual completed project volume = actual completed volume × 7850kg/m 3 ÷ 1000. 9.根据权利要求8所述的基于人工智能的土木工程造价管理方法，其特征在于：9. The civil engineering cost management method based on artificial intelligence according to claim 8 is characterized in that: 体积计算规则，包括：Volume calculation rules, including: 对于砌体构件，按照如下规则计算：For masonry components, the calculation is based on the following rules: 标准砖砌体：实际完成工程量＝实际完成体积×0.95，其中，0.95为考虑灰缝的折减系数；Standard brickwork: actual completed work volume = actual completed volume × 0.95, where 0.95 is the reduction factor for considering mortar joints; 砌块砌体：实际完成工程量＝实际完成体积÷(砌块体积×1.1)，得到砌块数量，其中1.1为灰缝系数；Block masonry: Actual completed work volume = actual completed volume ÷ (block volume × 1.1), to get the number of blocks, where 1.1 is the mortar joint coefficient; 对于模板工程，按照如下规则计算：For template projects, calculations are performed according to the following rules: 梁模板：实际完成工程量＝(梁底宽+2×梁高)×实际完成长度；Beam formwork: actual completed work quantity = (beam bottom width + 2 × beam height) × actual completed length; 板模板：实际完成工程量＝实际完成面积；Plate formwork: actual completed project quantity = actual completed area; 将同一分项工程编码下的所有构件工程量按照相同计量单位累加，得到该分项的实际完成工程量。Add up the quantities of all components under the same sub-project code using the same unit of measurement to obtain the actual completed quantity of the sub-project. 10.根据权利要求2至9任一所述的基于人工智能的土木工程造价管理方法，其特征在于：10. The civil engineering cost management method based on artificial intelligence according to any one of claims 2 to 9, characterized in that: S5，比对实际已完成工程量与BIM设计模型中对应时间节点的计划完成工程量，计算进度偏差率，包括：S5: Compare the actual completed project volume with the planned completed project volume at the corresponding time node in the BIM design model and calculate the progress deviation rate, including: 获取当前施工日期，从所述BIM设计模型的计划进度数据中提取对应时间节点的计划完成工程量；Obtain the current construction date and extract the planned completion amount of the corresponding time node from the planned progress data of the BIM design model; 将所述各分项工程的实际已完成工程量与对应的计划完成工程量进行比对，计算各分项工程的进度偏差率：Compare the actual completed work volume of each sub-project with the corresponding planned completed work volume, and calculate the progress deviation rate of each sub-project: 进度偏差率＝(实际已完成工程量-计划完成工程量)/计划完成工程量×100％；Schedule deviation rate = (actual completed work volume - planned completed work volume) / planned completed work volume × 100%; 根据各分项工程在总工程中的权重系数，计算加权平均进度偏差率，其中权重系数根据各分项工程的合同金额占比确定；Calculate the weighted average schedule deviation rate based on the weight coefficient of each sub-project in the overall project, where the weight coefficient is determined based on the proportion of the contract amount of each sub-project; 当某分项工程的进度偏差率超过预设的预警阈值时，生成进度预警信息，所述预警信息包括分项工程名称、偏差率数值和偏差原因分析；When the progress deviation rate of a sub-project exceeds the preset warning threshold, a progress warning message is generated, which includes the sub-project name, deviation rate value and deviation cause analysis; 根据进度偏差率的正负值判断工程进度状态：The project progress status is judged based on the positive and negative values of the progress deviation rate: 当进度偏差率为正值时，表示工程进度超前；When the progress deviation rate is positive, it means that the project is ahead of schedule; 当进度偏差率为负值时，表示工程进度滞后；When the progress deviation rate is negative, it means that the project progress is lagging behind; 生成进度偏差分析报表，包含各分项工程的计划工程量、实际工程量、偏差率和累计完成百分比。Generate a progress deviation analysis report, including the planned project quantity, actual project quantity, deviation rate and cumulative completion percentage of each sub-project.
CN202510972020.3A 2025-07-15 2025-07-15 An Artificial Intelligence-Based Method for Civil Engineering Cost Management Active CN120633942B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202510972020.3A CN120633942B (en) 2025-07-15 2025-07-15 An Artificial Intelligence-Based Method for Civil Engineering Cost Management

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202510972020.3A CN120633942B (en) 2025-07-15 2025-07-15 An Artificial Intelligence-Based Method for Civil Engineering Cost Management

Publications (2)

Publication Number Publication Date
CN120633942A true CN120633942A (en) 2025-09-12
CN120633942B CN120633942B (en) 2025-12-16

Family

ID=96965840

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202510972020.3A Active CN120633942B (en) 2025-07-15 2025-07-15 An Artificial Intelligence-Based Method for Civil Engineering Cost Management

Country Status (1)

Country Link
CN (1) CN120633942B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN121118225A (en) * 2025-11-17 2025-12-12 深圳大学 Real-model consistency comparison methods, devices, equipment, media and products
CN121544218A (en) * 2026-01-20 2026-02-17 中国电建集团贵阳勘测设计研究院有限公司 A method and apparatus for batch inspection of earthwork BIM models

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109118564A (en) * 2018-08-01 2019-01-01 湖南拓视觉信息技术有限公司 A kind of three-dimensional point cloud labeling method and device based on fusion voxel
CN115758536A (en) * 2022-11-22 2023-03-07 卓宇智能科技有限公司 Construction cost estimation method and estimation device based on three-dimensional scanning technology
CN118261748A (en) * 2024-03-21 2024-06-28 东南大学 A tunnel engineering progress and measurement linkage management method and system based on BIM technology
CN119229285A (en) * 2024-09-13 2024-12-31 苏州科技大学 A method for accurately extracting building facade structural lines based on laser point cloud feature depth
CN119339007A (en) * 2024-12-20 2025-01-21 杭州宇泛智能科技股份有限公司 Adaptive modeling method of scene space model based on multi-source data fusion
CN120124166A (en) * 2025-05-13 2025-06-10 中建五局第四建设有限公司 A dynamic monitoring method and system for deep foundation pit construction based on BIM

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109118564A (en) * 2018-08-01 2019-01-01 湖南拓视觉信息技术有限公司 A kind of three-dimensional point cloud labeling method and device based on fusion voxel
CN115758536A (en) * 2022-11-22 2023-03-07 卓宇智能科技有限公司 Construction cost estimation method and estimation device based on three-dimensional scanning technology
CN118261748A (en) * 2024-03-21 2024-06-28 东南大学 A tunnel engineering progress and measurement linkage management method and system based on BIM technology
CN119229285A (en) * 2024-09-13 2024-12-31 苏州科技大学 A method for accurately extracting building facade structural lines based on laser point cloud feature depth
CN119339007A (en) * 2024-12-20 2025-01-21 杭州宇泛智能科技股份有限公司 Adaptive modeling method of scene space model based on multi-source data fusion
CN120124166A (en) * 2025-05-13 2025-06-10 中建五局第四建设有限公司 A dynamic monitoring method and system for deep foundation pit construction based on BIM

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
BRAUN, A ET AL.: ""Towards automated construction progress monitoring using BIM-based point cloud processing"", 《EWORK AND EBUSINESS IN ARCHITECTURE, ENGINEERING AND CONSTRUCTION 2014》, 31 December 2015 (2015-12-31) *
莫鲁晟: ""基于BIM和三维激光扫描的桥梁钢筋施工监控方法研究"", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》, no. 05, 15 May 2024 (2024-05-15) *
齐慧慧: ""BIM协同多源数据建筑施工动态监测方法"", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》, no. 06, 15 June 2025 (2025-06-15), pages 42 - 63 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN121118225A (en) * 2025-11-17 2025-12-12 深圳大学 Real-model consistency comparison methods, devices, equipment, media and products
CN121544218A (en) * 2026-01-20 2026-02-17 中国电建集团贵阳勘测设计研究院有限公司 A method and apparatus for batch inspection of earthwork BIM models
CN121544218B (en) * 2026-01-20 2026-04-17 中国电建集团贵阳勘测设计研究院有限公司 Batch inspection method and device for earth-rock BIM model

Also Published As

Publication number Publication date
CN120633942B (en) 2025-12-16

Similar Documents

Publication Publication Date Title
CN120633942B (en) An Artificial Intelligence-Based Method for Civil Engineering Cost Management
CN108846888B (en) Automatic extraction method for fine size information of ancient wood building components
Biosca et al. Unsupervised robust planar segmentation of terrestrial laser scanner point clouds based on fuzzy clustering methods
Hichri et al. Review of the “as-buit BIM” approaches
Pătrăucean et al. State of research in automatic as-built modelling
CN112989453B (en) BIM-based holographic deformation information extraction method
CN117854060B (en) Tunnel rock body planar crack identification method and system based on deep learning
CN116883754A (en) Building information extraction method for ground LiDAR point cloud
CN114332291B (en) Method for extracting outline rule of oblique photography model building
Lin et al. Scan vs. BIM: Automated geometry detection and BIM updating of steel framing through laser scanning
CN105184854B (en) For the fast modeling method of underground space scanning element cloud performance data
CN118762051A (en) A regularity-constrained step line extraction algorithm for open-pit mines
CN121213555B (en) Unmanned plane laser radar-based building point cloud deformation quantitative extraction method
CN120931145A (en) Building decoration quality analysis system based on BIM technology
CN120429934A (en) Steel structure installation adjustment method and system based on dynamic tracking of column axis
Liu et al. Rigorous density correction model for single-scan TLS point clouds
CN119559241A (en) LiDAR granary volume calculation method and system based on material surface point cloud compensation
CN101071453A (en) System and method for using 3D scanning data recognizing initial design intention
Liang et al. Extraction of Feature Information from Point Cloud of Large Volume Steel Truss Members
CN121207065B (en) A method and apparatus for measuring the overall large deformation of tunnels based on laser scanning
CN119850689B (en) A component deviation measurement method, device, system and storage medium
CN118897528B (en) A workpiece measurement point sampling offline planning method and system
Zhang et al. Segmentation and modeling method of building structure point cloud based on BIM model
CN112465896B (en) A method for detecting geometric changes of 3D murals based on multi-view contour points
Cheng et al. TOWARDS AUTOMATED GENERATION OF PARAMETRIC BIM FOR STEEL STRUCTURES BASED ON LASER SCANNING DATA

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
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20260129

Address after: 528200, 1014, 1015, 1, Nanhai digital new town, Nanhai Road, Nanhai District, Guangdong, Guicheng, 1, Foshan

Patentee after: Guangdong Chong Nan Engineering Management Co.,Ltd.

Country or region after: China

Address before: 230092 Anhui Province, Hefei City, Baohe District, Hangzhou Road 3333, Hesheng Gongguan Building 9, Room 1003

Patentee before: Hefei Yinyue Space Architectural Design Co.,Ltd.

Country or region before: China