CN-121982022-A - Fan foundation steel bar quantity statistical method and system based on specification constraint

CN121982022ACN 121982022 ACN121982022 ACN 121982022ACN-121982022-A

Abstract

The application discloses a method and a system for counting the number of reinforcing steel bars of a fan foundation based on specification constraint, and relates to the field of intelligent accounting of engineering quantity; the method comprises the steps of carrying out registration denoising on point cloud based on reference control points to obtain clean point cloud, extracting section characteristics along highly layered slices and detecting to obtain candidate sections, constructing specification constraint factors according to theoretical specification parameters, grading and screening the candidate sections to obtain effective sections, carrying out space consistency check and anomaly correction by combining arrangement rules and design tolerances, counting the number of reinforcing steel bars of each specification, and outputting a statistical report. Therefore, high-precision calculation quantity which can be rechecked according to the specification can be realized under complex field conditions such as shielding and overlapping.

Inventors

GAO BIN
JIA WENHUI
MENG XIANGXIN
CHEN SHILIANG
ZHOU AIGUO
Qu Fupeng
WU XIAOYONG
HUO YI

Assignees

五环绿能(北京)工程科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260402

Claims (8)

1. The method for counting the number of the fan foundation steel bars based on the specification constraint is characterized by comprising the following steps: Acquiring design construction data of a fan foundation and scanning point cloud data of a construction site, wherein the design construction data comprises theoretical specification parameters and arrangement rules of steel bars, and the scanning point cloud data reflects the three-dimensional space form of a steel bar cage; performing coordinate system registration and denoising preprocessing on the scanning point cloud data based on reference control points in the design construction data to generate net point cloud data consistent with the design construction data; Carrying out layering slicing processing on the net point cloud data along the height direction, generating slicing characteristic data reflecting the section distribution of the reinforcing steel bars at different height layers, and carrying out target detection on each slicing characteristic data to obtain an initial section candidate set; Constructing a specification constraint factor by utilizing the theoretical specification parameters, and evaluating and screening the confidence coefficient of the initial section candidate set based on the specification constraint factor to identify an effective steel bar section conforming to the design specification, wherein the specification constraint factor is used for representing the coincidence degree of a candidate target and the theoretical specification parameters; And based on the arrangement rules and the design tolerance ranges, carrying out space consistency verification and anomaly correction on the effective steel bar sections, counting the number of steel bars of each specification according to the corrected steel bar section detection results, and outputting a steel bar number counting report.
2. The method of claim 1, wherein the performing coordinate system registration and denoising preprocessing on the scan point cloud data based on the reference control points in the design configuration data to generate net point cloud data consistent with the design configuration data coordinates comprises: Firstly extracting basic geometric center and edge contour features in the scan point cloud data, aligning the basic geometric center and edge contour features with reference control points in the design construction data to finish coarse registration, and then minimizing a space distance error between the point cloud and the design model by using an iterative optimization algorithm to solve optimal space transformation parameters so as to accurately unify the scan point cloud data to the design coordinate system; And based on the partial geometrical feature analysis of the point cloud, calculating the normal vector and the partial curvature of each point in the rest point cloud, identifying a planar point cloud cluster with the consistent normal vector direction and the partial curvature lower than a smoothness threshold, judging the planar point cloud cluster as the interference of a concrete pile casing surface or a construction template, removing the interference, and reserving the linear point cloud cluster with the high curvature characteristic as the net point cloud data for representing the reinforcement cage.
3. The method of claim 1, wherein the hierarchically slicing the net point cloud data in the height direction to generate slice feature data reflecting a profile of the reinforcing steel bar at different height layers, and performing target detection for each slice feature data to obtain an initial cross-section candidate set, comprises: setting the slice thickness according to the minimum steel bar diameter in the theoretical specification parameters so as to ensure that slice data can contain complete steel bar section information; constructing a two-dimensional projection grid for each slice, counting the number of point clouds falling into each grid unit, normalizing, and generating a gray image reflecting the distribution density of the point clouds of the slice; according to the minimum steel bar diameter self-adaptive construction morphological structure elements, performing morphological edge enhancement operation on the gray level image by utilizing the structure elements, and calculating the difference between a morphological expansion result and a morphological corrosion result to highlight the edge profile of the steel bar section, so as to generate a morphological gradient feature map as the slice feature data; And taking the morphological gradient feature map as input to execute Hough circle transformation, positioning a high response area by utilizing an accumulated voting mechanism of gradient amplitude values, and extracting circle centers and radiuses of a plurality of circular targets to form the initial section candidate set.
4. The method of claim 1, wherein constructing a specification constraint factor using the theoretical specification parameter and performing confidence evaluation and screening on the initial cross-section candidate set based on the specification constraint factor to identify effective rebar cross-sections that meet a design specification comprises: analyzing design construction data corresponding to the current slicing level, and extracting theoretical design radius and theoretical design adjacent distance of the level steel bar as prior constraint parameters; Constructing a specification constraint factor comprising geometric consistency evaluation and space topology consistency evaluation aiming at any one candidate target in the initial section candidate set, wherein the specification constraint factor is configured to utilize exponential decay function logic to reduce the weight of the candidate target when the detection radius of the candidate target deviates from the theoretical design radius or the distance between the candidate target and a neighborhood target deviates from the theoretical design adjacent distance; The visual characteristics and the specification constraint are fused, and the comprehensive confidence score of each candidate target is calculated, wherein the comprehensive confidence score is formed by the weighted sum of the visual gradient response intensity of the candidate target in the slice characteristic data and the specification constraint factor; And performing iterative screening on the initial section candidate set based on the comprehensive confidence score, wherein the target with the highest score is preferentially reserved, and redundant targets with excessively high overlapping rate are restrained until all candidate targets are traversed, so that the effective steel bar sections meeting the design specification are output.
5. The method of claim 1, wherein the performing spatial consistency check and anomaly correction on the effective rebar section based on the arrangement rules and design tolerance ranges comprises: Classifying and dividing the effective steel bar sections into corresponding design specification sets by utilizing a nearest neighbor matching principle according to the arrangement rules; Based on the classification result, constructing a global specification error evaluation model, and quantifying the sum of geometric deviation and topological distribution deviation of all effective sections in a current level relative to an ideal design model in a weighted aggregation mode, wherein the geometric deviation represents the difference between a detection radius and a theoretical radius, the topological distribution deviation represents the difference between an actual distance and a theoretical distance, and the model comprises a weight item for balancing the numerical magnitude difference between the radius and the distance; Calculating a global specification error evaluation value of the global specification error evaluation model, and comparing the global specification error evaluation value with a tolerance threshold set based on a design tolerance range; If the global specification error evaluation value exceeds the tolerance threshold, judging that the detection result of the current level has systematic deviation or local shielding abnormality, and triggering to execute abnormality correction operation.
6. The method of claim 5, wherein the anomaly correction operation comprises a miss-complement operation for an occlusion region, comprising: Traversing an effective steel bar section set in the current level, analyzing a space topological relation between adjacent sections, and calculating the actual physical distance of the space topological relation; If the actual physical distance between the adjacent sections is detected to be larger than the theoretical design distance and the area is a rib distribution area in the design and construction data, judging that the steel bars are missing due to shielding; Based on the space coordinates of the known effective steel bar sections at the two ends of the adjacent sections and the theoretical design distance, deducing the theoretical number of the missing steel bars; And generating a corresponding number of virtual steel bar section center coordinates between the known coordinates at the two ends of the adjacent sections according to a theoretical design interval by utilizing linear spatial interpolation logic, and merging the generated virtual steel bar sections into the effective steel bar section set, thereby obtaining the corrected steel bar section detection result.
7. The method of claim 1, wherein counting the number of the steel bars of each specification according to the corrected steel bar section detection result and outputting a steel bar number counting report, comprises: aggregating the corrected detection results of the sections of the steel bars according to the specification types of the steel bars, counting the total detection quantity of algorithms of various specifications, and calling the total quantity of corresponding theoretical designs; defining allowable deviation threshold values for various specifications, and constructing a segment arbitration decision logic to determine the final output statistical quantity: When the difference value between the total number detected by the algorithm and the total number of theoretical design is within the allowable deviation threshold value range, judging that the detection result is effective, and adopting the total number detected by the algorithm as an output result; When the difference exceeds the allowable deviation threshold, judging that the detection result has serious deviation, and adopting the theoretical design total number as an output result; and triggering an abnormal warning in the steel bar quantity statistics report for the condition that the difference value exceeds the limit, marking the position of an abnormal region and associating the point cloud visualization snapshot of the region for manual review.
8. A fan foundation rebar quantity statistics system based on specification constraints, the system comprising: The system comprises a data acquisition unit, a control unit and a control unit, wherein the data acquisition unit is used for acquiring design construction data of a fan foundation and scanning point cloud data of a construction site, the design construction data comprise theoretical specification parameters and arrangement rules of steel bars, and the scanning point cloud data reflect three-dimensional space forms of a steel bar cage; the data registration and preprocessing unit is used for carrying out coordinate system registration and denoising preprocessing on the scanning point cloud data based on the reference control points in the design construction data to generate net point cloud data consistent with the design construction data; The slice and target detection unit is used for carrying out layered slice processing on the net point cloud data along the height direction, generating slice characteristic data reflecting the section distribution of the reinforcing steel bars at different heights, and carrying out target detection on each slice characteristic data to obtain an initial section candidate set; the specification constraint and screening unit is used for constructing a specification constraint factor by utilizing the theoretical specification parameters, and evaluating and screening the confidence coefficient of the initial section candidate set based on the specification constraint factor so as to identify an effective steel bar section conforming to the design specification, wherein the specification constraint factor is used for representing the coincidence degree of the candidate target and the theoretical specification parameters; And the verification and statistics report unit is used for carrying out space consistency verification and abnormality correction on the effective steel bar section based on the arrangement rule and the design tolerance range, counting the number of steel bars of each specification according to the corrected steel bar section detection result, and outputting a steel bar number statistics report.

Description

Fan foundation steel bar quantity statistical method and system based on specification constraint Technical Field The application relates to the technical field of intelligent accounting of engineering quantities, in particular to a fan foundation steel bar quantity counting method and system based on specification constraint. Background In the field of wind power engineering construction, a fan foundation is used as a key structure for bearing huge load, a large-diameter concrete round table or cylinder type design is generally adopted, and the arrangement of internal reinforcing steel bars has the remarkable characteristics of high density, multi-specification intersection and layered arrangement in large-volume concrete. The accurate statistics of engineering quantity, especially the accounting of the number of reinforcing steel bars, is directly related to the cost control, the structural safety assessment and the acceptance of construction standardization of projects. However, due to the specificity and complexity of the fan infrastructure, conventional engineering quantity statistics means have been difficult to meet the requirements of modern fine construction. At present, the steel bar statistical method is mainly divided into manual calculation, a two-dimensional graphic method and automatic calculation based on general BIM software. The manual and two-dimensional methods rely on mechanical turning and experience estimation of the design drawing, and omission or miscalculation is very easy to occur and the efficiency is low when facing the irregular three-dimensional special-shaped reinforcement structure of the fan foundation. Although some projects introduce BIM technology, the existing BIM calculation amount often depends on a standardized component Family (Family), and complex steel bar bending, staggering and variable-section arrangement in a fan foundation are difficult to flexibly adapt, so that the generated detail table often cannot truly reflect the actual space parameters of each steel bar. In addition, with the development of machine vision technology, some reinforcing steel bar counting schemes based on image recognition appear in the industry, but the schemes are limited to reinforcing steel bar processing fields or stockpiling areas, and are aimed at discrete reinforcing steel bars with clearly visible strip-shaped sections. In a fan foundation construction site, steel bars are usually bound into a cage, and part of the steel bars are covered by a template or concrete, so that a visual section is extremely limited and is accompanied by serious shielding and overlapping, and the current visual analysis system is difficult to realize high-precision quantity statistics and compliance verification in a complex shielding environment. Disclosure of Invention The application provides a fan foundation reinforcement quantity counting method, a system, a storage medium, a computer program product and electronic equipment based on specification constraint, which are used for at least solving the problem that in the prior art, high-precision and checkable automatic statistics of fan foundation reinforcement engineering quantity statistics are difficult to realize under the conditions of complex three-dimensional structure and on-site shielding interference. According to the method, design construction data of a fan foundation and scanning point cloud data of a construction site are obtained, the design construction data comprise theoretical specification parameters of steel bars and arrangement rules, the scanning point cloud data reflect three-dimensional space forms of reinforcement cages, coordinate system registration and denoising preprocessing are conducted on the scanning point cloud data based on reference control points in the design construction data to generate net point cloud data consistent with the coordinates of the design construction data, hierarchical slicing is conducted on the net point cloud data in the height direction to generate slice feature data reflecting section distribution of the steel bars at different levels, target detection is conducted on each slice feature data to obtain an initial section candidate set, confidence evaluation and screening are conducted on the initial section candidate set based on the theoretical specification parameters to identify effective sections conforming to design specifications, the constraint factors are used for representing the matching degree of the candidate targets and the theoretical specification parameters, correction results of the steel bars are reported according to the space correction results of the steel bars and the correction results of the steel bar statistics. In a second aspect, the embodiment of the application provides a fan foundation steel bar quantity statistical system based on specification constraint, which comprises a data acquisition unit, a data registration and preprocessing unit, a specification const