CN117951546B - BIM-combined concrete structure defect detection method and system - Google Patents
BIM-combined concrete structure defect detection method and system Download PDFInfo
- Publication number
- CN117951546B CN117951546B CN202410350098.7A CN202410350098A CN117951546B CN 117951546 B CN117951546 B CN 117951546B CN 202410350098 A CN202410350098 A CN 202410350098A CN 117951546 B CN117951546 B CN 117951546B
- Authority
- CN
- China
- Prior art keywords
- bim
- concrete
- data
- target
- defect
- 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.)
- Active
Links
- 230000007547 defect Effects 0.000 title claims abstract description 268
- 238000001514 detection method Methods 0.000 title claims abstract description 182
- 238000000034 method Methods 0.000 claims abstract description 94
- 238000005065 mining Methods 0.000 claims abstract description 71
- 238000000547 structure data Methods 0.000 claims description 285
- 238000012549 training Methods 0.000 claims description 169
- 239000013598 vector Substances 0.000 claims description 116
- 230000007847 structural defect Effects 0.000 claims description 80
- 238000010586 diagram Methods 0.000 claims description 65
- 238000012545 processing Methods 0.000 claims description 45
- 238000004458 analytical method Methods 0.000 claims description 32
- 238000002372 labelling Methods 0.000 claims description 19
- 238000001212 derivatisation Methods 0.000 claims description 9
- 238000013139 quantization Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 8
- 238000012423 maintenance Methods 0.000 abstract description 52
- 230000008569 process Effects 0.000 abstract description 44
- 238000011156 evaluation Methods 0.000 abstract description 9
- 238000013135 deep learning Methods 0.000 abstract description 7
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 239000000463 material Substances 0.000 description 21
- 238000010276 construction Methods 0.000 description 12
- 238000005457 optimization Methods 0.000 description 8
- 239000000284 extract Substances 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 230000008439 repair process Effects 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 230000003631 expected effect Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011049 filling Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000000246 remedial effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012916 structural analysis Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/38—Concrete; Lime; Mortar; Gypsum; Bricks; Ceramics; Glass
- G01N33/383—Concrete or cement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2132—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on discrimination criteria, e.g. discriminant analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
- G01N2021/8861—Determining coordinates of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Artificial Intelligence (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Analytical Chemistry (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Geometry (AREA)
- Biochemistry (AREA)
- Computer Hardware Design (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Food Science & Technology (AREA)
- Software Systems (AREA)
- Ceramic Engineering (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Medicinal Chemistry (AREA)
- Computing Systems (AREA)
- Structural Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Signal Processing (AREA)
Abstract
The embodiment of the invention relates to the technical field of artificial intelligence, in particular to a concrete structure defect detection method and system combined with BIM, which are high in efficiency, accuracy and strong in adaptability by combining BIM structure attribute mining, cloud data retrieval, structure defect matching feature sets and deep learning technology. The method not only improves the efficiency and accuracy of defect detection, but also can process various complex and diversified BIM data structures, and provides powerful technical support for safety evaluation and maintenance of the concrete structure.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a concrete structure defect detection method and system combined with BIM.
Background
In the field of concrete structure defect detection, with the wide application of Building Information Model (BIM) technology, BIM structure data has become an important basis for evaluating the structural integrity and safety of concrete. Traditional concrete structure defect detection methods often rely on manual inspection and local nondestructive testing, and the methods are low in efficiency and difficult to comprehensively and accurately identify potential defects in the structure. It is therefore important to develop a system and method that automatically, efficiently and accurately detects defects in concrete structures.
In recent years, with the development of computer vision and deep learning technology, an intelligent structural defect detection method based on BIM data gradually becomes a research hotspot. These methods typically involve automatic analysis of BIM structure data to extract critical structural attribute information and use the data to identify potential structural defects. However, existing methods still face challenges in handling complex concrete structures, such as efficiency of data processing, accuracy of defect identification, and adaptability to different types of BIM data.
Disclosure of Invention
In order to improve the technical problems in the related art, the invention provides a concrete structure defect detection method and system combined with BIM.
In a first aspect, an embodiment of the present invention provides a concrete structure defect detection method combined with BIM, which is applied to a concrete structure defect detection system, and the method includes:
Acquiring target concrete BIM structure data currently transmitted by a detection interface of a target task server in the concrete structure defect detection system;
Performing BIM structure attribute mining on the target concrete BIM structure data to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data;
Determining associated concrete BIM structure data corresponding to the target concrete BIM structure data from initial concrete BIM structure data of a cloud BIM structure information set according to the target BIM structure attribute vector;
When the commonality index between the associated concrete BIM structure data and the target concrete BIM structure data accords with a set commonality requirement, a structure defect matching feature set is obtained, wherein the structure defect matching feature set comprises defect state matching features between initial concrete BIM structure data and set defect labeling data; determining a target structural defect detection result corresponding to the associated concrete BIM structure data based on the structural defect matching feature set and the associated concrete BIM structure data, and issuing the target structural defect detection result for the target concrete BIM structure data to the target task server;
When the commonality index between the associated concrete BIM structure data and the target concrete BIM structure data does not meet the set commonality requirement, carrying out structure defect judging operation on the target concrete BIM structure data through a defect judging network to obtain a structure defect detection result corresponding to the target concrete BIM structure data, and issuing the structure defect detection result for the target concrete BIM structure data to the target task server;
With reference to the first aspect, in a possible implementation manner of the first aspect, the initial concrete BIM structure data includes a BIM element unit and a concrete space structure diagram, and the set defect labeling data includes element trend characteristics and mechanical conduction data; before the obtaining the structural defect matching feature set, the method further comprises:
Constructing a first defect state matching feature between the BIM element unit and the element trend feature thereof;
Determining mechanical conduction data for a BIM element unit, and constructing a second defect state matching feature between the BIM element unit and the mechanical conduction data;
constructing a third defect state matching characteristic between a concrete space structure diagram and mechanical conduction data corresponding to the concrete space structure diagram;
Constructing a fourth defect state matching feature between the concrete space structure diagram and the corresponding element trend feature;
A structural defect matching feature set is determined based on the first defect state matching feature, the second defect state matching feature, the third defect state matching feature, and the fourth defect state matching feature.
With reference to the first aspect, in a possible implementation manner of the first aspect, the determining, according to the target BIM structure attribute vector, associated concrete BIM structure data corresponding to the target concrete BIM structure data from initial concrete BIM structure data of a cloud BIM structure information set includes:
Identifying a building scene tag to which the target concrete BIM structure data belongs;
Determining initial concrete BIM structure data under the building scene label in the cloud BIM structure information set;
And determining associated concrete BIM structure data corresponding to the target concrete BIM structure data from the initial concrete BIM structure data under the building scene label according to the target BIM structure attribute vector.
With reference to the first aspect, in a possible implementation manner of the first aspect, the determining, based on the structural defect matching feature set and the associated concrete BIM structural data, a target structural defect detection result corresponding to the associated concrete BIM structural data includes:
Determining target set defect labeling data corresponding to the associated concrete BIM structure data from the structure defect matching feature set;
When the target set defect labeling data comprises mechanical conduction data aiming at the associated concrete BIM structure data, processing the target concrete BIM structure data according to the mechanical conduction data to obtain a target structure defect detection result.
With reference to the first aspect, in a possible implementation manner of the first aspect, the obtaining target concrete BIM structure data currently input by a detection interface of the target task server in the concrete structure defect detection system includes:
Generating at least one template concrete BIM structure data according to the building scene label of the target task server, and transmitting the template concrete BIM structure data to the target task server in a detection interface in the concrete structure defect detection system;
And selecting and processing the transmitted template concrete BIM structure data in the detection interface according to the target task server to acquire the target concrete BIM structure data currently transmitted by the target task server in the detection interface.
With reference to the first aspect, in a possible implementation manner of the first aspect, after the obtaining, by the target task server, target concrete BIM structure data currently input by a detection interface in the concrete structure defect detection system, the method further includes:
Carrying out concrete BIM structure data quality analysis on the target concrete BIM structure data;
when the quality analysis report accords with a set index, transmitting concrete BIM structure data under a template BIM element to the target task server, wherein the template BIM element is different from BIM elements contained in the target concrete BIM structure data;
And when the quality analysis report does not accord with the set index, transmitting basic concrete BIM structure data under a target BIM element to the target task server, wherein the target BIM element is associated with BIM elements contained in the target concrete BIM structure data.
With reference to the first aspect, in one possible implementation manner of the first aspect, the performing BIM structure attribute mining on the target concrete BIM structure data to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data includes:
And performing BIM structure attribute mining on the target concrete BIM structure data through a BIM structure attribute mining algorithm to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data.
With reference to the first aspect, in one possible implementation manner of the first aspect, before performing BIM structure attribute mining on the target concrete BIM structure data by using a BIM structure attribute mining algorithm to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data, the method further includes:
Acquiring an algorithm sample, wherein the algorithm sample comprises a plurality of concrete BIM structure data examples;
Generating, for each concrete BIM structure data instance, a set of positive example tuples of the concrete BIM structure data instance, the set of positive example tuples comprising the concrete BIM structure data instance and at least one positive training data having a feature in common with the concrete BIM structure data instance;
sorting the concrete BIM structure data examples with each positive training data respectively to construct at least one first training example binary set of the concrete BIM structure data examples;
constructing at least one second training example binary group of the concrete BIM structure data example according to the concrete BIM structure data example and data in the positive example binary group set of each concrete BIM structure data example;
Performing BIM structure attribute mining on the data in the first training example binary group and the second training example binary group through a preset BIM structure attribute mining algorithm to obtain BIM structure attribute vectors of the data in the first training example binary group and the second training example binary group;
Determining attribute vector differences of the first training example binary group based on BIM structure attribute vectors of each content in the first training example binary group;
Determining attribute vector differences of the second training example binary group based on BIM structure attribute vectors of contents in the second training example binary group;
and optimizing algorithm variables of a preset BIM structure attribute mining algorithm according to the attribute vector difference of the first training example binary group and the attribute vector difference of the second training example binary group to obtain an optimized BIM structure attribute mining algorithm.
With reference to the first aspect, in one possible implementation manner of the first aspect, the concrete BIM structure data example includes a BIM element unit and a concrete space structure diagram example; the generating the positive example binary set of concrete BIM structure data examples includes:
Respectively carrying out image data derivatization on the BIM element unit and the concrete space structure diagram example to obtain a derivatization element unit of the BIM element unit and a concrete space structure derivatization diagram of the concrete space structure diagram example;
Determining a background image in the concrete space structure diagram example based on the set element word vector and a preset BIM element unit; updating the background image in the concrete space structure diagram example to obtain an updated concrete space structure diagram example;
Acquiring quantization characteristics of the BIM element unit; determining positive training data of the BIM element unit according to the derivative element unit and the quantization characteristic of the BIM element unit; and determining positive training data of the concrete space structure diagram example according to the concrete space structure derivative diagram of the concrete space structure diagram example and the updated concrete space structure diagram example.
With reference to the first aspect, in one possible implementation manner of the first aspect, the optimizing an algorithm variable of a preset BIM structure attribute mining algorithm according to the attribute vector difference of the first training example binary set and the attribute vector difference of the second training example binary set to obtain an optimized BIM structure attribute mining algorithm includes:
Integrating the attribute vector differences of each second training example binary group to obtain integrated second attribute vector differences;
for each first training example binary group, determining local training cost information of the first training example binary group according to the integrated second attribute vector difference and the attribute vector difference of the first training example binary group;
Integrating the local training cost information of each first training example binary group to obtain global training cost information;
and optimizing algorithm variables of a preset BIM structure attribute mining algorithm according to the global training cost information to obtain an optimized BIM structure attribute mining algorithm.
In a second aspect, the present invention also provides a system for detecting defects of a concrete structure, including: a memory for storing program instructions and data; and a processor coupled to the memory for executing instructions in the memory to implement the method as described above.
In a third aspect, the present invention also provides a computer storage medium containing instructions which, when executed on a processor, implement the above-described method.
Aiming at the problems, the invention provides a concrete structure defect detection method based on BIM structure attribute mining and deep learning technologies. The method comprises the steps of firstly, obtaining target concrete BIM structure data currently transmitted by a detection interface of a concrete structure defect detection system in a target task server. Then, by performing BIM structure attribute mining on the data, key structure attribute information is extracted and converted into BIM structure attribute vectors. These vectors are then used to retrieve BIM data associated with the target concrete BIM structure data from the cloud BIM structure information set.
After retrieving the associated concrete BIM structure data, the present invention further evaluates the commonality index between these data and the target concrete BIM structure data. If the commonality index meets the set requirement, the correlation data and the target data are indicated to have higher similarity in structure, and at the moment, the structural defect detection result corresponding to the correlation data can be determined by utilizing the pre-established structural defect matching characteristic set. The feature set comprises matching features between the initial concrete BIM structure data and known defect labeling data, and provides important reference information for defect detection.
However, if the commonality index does not meet the set requirement, it indicates that there is a large difference between the associated data and the target data in structure, and at this time, defect detection cannot be performed directly by using the structural defect matching feature set. Under the condition, the invention adopts a defect judging network to directly judge the structural defect of the target concrete BIM structural data. The network is trained, so that various structural defects in BIM data can be identified, and corresponding detection results are output.
In summary, the concrete structure defect detection method is efficient, accurate and high in adaptability by combining BIM structure attribute mining, cloud data retrieval, structure defect matching feature sets and deep learning technology. The method not only improves the efficiency and accuracy of defect detection, but also can process various complex and diversified BIM data structures, and provides powerful technical support for safety evaluation and maintenance of the concrete structure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
Fig. 1 is a schematic flow chart of a concrete structure defect detection method combined with BIM according to an embodiment of the present invention.
Fig. 2 is a block diagram of a concrete structure defect detection system 300 according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention will be described below with reference to the accompanying drawings.
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the invention. Rather, they are merely examples of apparatus and methods consistent with aspects of the invention.
It should be noted that the terms "first," "second," and the like in the description of the present invention and the above-described drawings are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order.
The method embodiment provided by the embodiment of the invention can be executed in a concrete structure defect detection system, computer equipment or similar computing devices. Taking as an example operation on a concrete structure defect detection system, the concrete structure defect detection system may comprise one or more processors (which may include, but is not limited to, a microprocessor MCU or a processing device such as a programmable logic device FPGA) and a memory for storing data, and optionally the concrete structure defect detection system may further comprise a transmission device for communication functions. It will be appreciated by those skilled in the art that the above-described structure is merely illustrative and is not intended to limit the structure of the above-described concrete structure defect detection system. For example, the concrete structure defect detection system may also include more or fewer components than those shown above, or have a different configuration than those shown above.
The memory may be used to store a computer program, for example, a software program of application software and a module, for example, a computer program corresponding to a concrete structure defect detection method combined with BIM in the embodiment of the present invention, and the processor executes the computer program stored in the memory, thereby performing various functional applications and data processing, that is, implementing the method. The memory may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid state memory. In some examples, the memory may further include memory remotely located with respect to the processor, the remote memory being connectable to the concrete structure defect detection system through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission means is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the concrete structure defect detection system. In one example, the transmission means comprises a network adapter (Network Interface Controller, simply referred to as NIC) that can be connected to other network devices via a base station to communicate with the internet. In one example, the transmission device may be a Radio Frequency (RF) module, which is used to communicate with the internet wirelessly.
Referring to fig. 1, fig. 1 is a schematic flow chart of a concrete structure defect detection method combined with BIM according to an embodiment of the present invention, where the method is applied to a concrete structure defect detection system, and further includes steps 110 to 150.
And 110, acquiring target concrete BIM structure data currently transmitted by a detection interface of a target task server in the concrete structure defect detection system.
And 120, performing BIM structure attribute mining on the target concrete BIM structure data to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data.
And 130, determining associated concrete BIM structure data corresponding to the target concrete BIM structure data from initial concrete BIM structure data of a cloud BIM structure information set according to the target BIM structure attribute vector.
Step 140, when the commonality index between the associated concrete BIM structure data and the target concrete BIM structure data accords with a set commonality requirement, a structure defect matching feature set is obtained, wherein the structure defect matching feature set comprises defect state matching features between initial concrete BIM structure data and set defect marking data; and determining a target structural defect detection result corresponding to the associated concrete BIM structure data based on the structural defect matching feature set and the associated concrete BIM structure data, and issuing the target structural defect detection result for the target concrete BIM structure data to the target task server.
And 150, when the commonality index between the associated concrete BIM structure data and the target concrete BIM structure data does not meet the set commonality requirement, performing structural defect distinguishing operation on the target concrete BIM structure data through a defect distinguishing network to obtain a structural defect detection result corresponding to the target concrete BIM structure data, and issuing the structural defect detection result corresponding to the target concrete BIM structure data to the target task server.
In embodiments of the present invention, a concrete structure defect detection system is performing a critical task that involves performing detailed defect detection on the concrete structure of a building. The following is a detailed description of how the system accomplishes this task step by step.
First, the system obtains a specific concrete BIM (Building Information Modeling, building information model) structure data incoming to the target task server. The data is detailed three-dimensional information about a specific part of the building, including its geometry, physical characteristics, construction details, etc.
Next, the system performs deep mining on the segment of BIM structure data, extracts a series of BIM structure attributes inherent therein, and organizes the attributes into an attribute vector. This vector, like a "digital fingerprint" of the building part, uniquely identifies its characteristics in terms of structure, materials, construction, etc.
The system then compares this "digital fingerprint" to the vast BIM structure information set stored in the cloud. The information set contains a large amount of known, validated concrete BIM structure data. Through an efficient algorithm, the system quickly finds out a few pieces of associated concrete BIM structure data which are most matched with the target BIM structure attribute vector.
At this time, the system makes a key judgment: whether the commonality index between the associated data and the target data meets the preset standard or not. The commonality index is a quantitative index for measuring the similarity of the two, and comprehensively considers factors of multiple dimensions, such as structural shape, material type, construction condition and the like.
If the commonality index meets the requirements, indicating that there is a high degree of structural similarity between the associated data and the target data, then the types of defects they may have are also most likely to be similar. The system then invokes a pre-established structural defect matching feature set that contains a number of known correspondences between concrete BIM structural data and various defect states. Based on the feature set and the associated data, the system can accurately predict the structural defect condition corresponding to the target data and immediately feed back the result to the target task server.
However, if the commonality index does not meet the requirement, which indicates that there is a large difference in structure between the associated data and the target data, then the defect condition of the target data cannot be predicted simply by comparison. In this case, the system starts a more complex defect discrimination network, and the network can directly perform detailed structural analysis on the target data through advanced technologies such as deep learning, so as to accurately identify various possible defects. Once the discrimination is completed, the system also rapidly feeds back the result to the target task server.
Through the process, the concrete structure defect detection system can effectively utilize BIM technology and big data analysis to perform comprehensive and high-precision defect detection on the concrete structure of the building, and provides powerful technical support for guaranteeing the safety of the building.
The relative terms in step 110, step 120 and step 130 are explained as follows.
A target task server refers to a server in a certain network or system that is dedicated to processing, storing, or forwarding a particular task or data. In a concrete structure defect detection system, it may be a central processing unit responsible for receiving, processing and returning the detection results. For example, when a concrete structure defect detection system needs to detect defects on a BIM model of a building, it will send tasks to a target task server. The server receives the model data, processes and analyzes the model data, and finally returns a detection result.
The detection interface is a standardized channel between systems or software components for transmitting data or instructions. In a concrete structure defect detection system, the detection interface may be a specific data input point for receiving BIM structure data to be detected. For example, a concrete structure defect detection system may provide an API (application program interface) as a detection interface, allowing other systems or software to directly transfer BIM data to the detection system for processing.
The target concrete BIM structure data refers to BIM model data of a specific portion where defect detection is required. These data contain geometric, physical and functional information of the concrete structure. For example, if a certain concrete beam section of a bridge is to be subjected to defect detection, the BIM model data of the beam section is target concrete BIM structure data.
BIM structure attribute mining refers to the process of extracting and analyzing structure attributes from BIM models. These properties may include material type, size, weight, connection, etc., which are critical to defect detection. For example, during BIM structure attribute mining, the system may analyze the diameter, height, concrete strength, etc. of a concrete column and use this attribute information for subsequent defect detection analysis.
The target BIM structure attribute vector is obtained by mining BIM structure attributes and is used for representing a set of numerical values or symbols of the structure characteristics of the BIM model. This vector is an important input to the defect detection algorithm. For example, for a particular concrete slab, the BIM structure attribute vector may include numerical information such as the thickness, area, number of support points, etc. of the slab. This vector will be used to compare to known BIM structure information to identify potential defects.
The cloud BIM structure information set refers to a large number of BIM (building information model) structure data sets stored on a cloud server. The data contains detailed three-dimensional information of various building structures that can be accessed and used by multiple users or systems simultaneously. Cloud storage makes these data easy to share, update, and maintain. For example, a building design company may upload BIM model data for all its items to a cloud BIM structure information set so that designers, engineers and clients of different departments can access the latest and most accurate building information at any time. In a concrete structure defect detection system, this set of information can be used as a reference database to assist the system in identifying and analyzing potential defects of the target structure.
The initial concrete BIM structure data refers to raw BIM model data that is already present before the concrete structure defect detection system begins to operate. These data are directly derived from the BIM model of the building project, reflecting the original design or construction state of the concrete structure. For example, when performing defect detection of a new building project, the detection system first needs to acquire BIM model data of the project as initial concrete BIM structure data. The data includes information about the geometry, dimensions, location, and material properties of concrete components such as beams, plates, columns, etc.
The associated concrete BIM structure data refers to BIM model data having similarity or correlation in structure, shape, material or other characteristics with the target concrete BIM structure data in the concrete structure defect detection process. Such data may be from a cloud BIM structure information set or other known BIM databases. For example, when the system detects defects on target concrete BIM structure data, it may search for several pieces of associated concrete BIM structure data similar to the target data in the cloud BIM structure information set. These associated data may come from the same type of building, components using the same construction method, or parts having similar geometries, etc. By comparing the commonalities and differences between the target data and the associated data, the system can more accurately identify the types of defects that may be present in the target data.
The relevant nouns in step 140 are explained as follows.
The commonality index is a quantization index used to measure the degree of similarity or commonality between two or more objects. In a concrete structure defect detection system, a commonality index is typically used to evaluate the similarity between target concrete BIM structure data and associated concrete BIM structure data. For example, the system may calculate the geometric similarity, material property consistency, etc. between the target concrete BIM data and the associated data, and combine these factors to give a commonality index. For example, if both are very similar in shape and material, the commonality index is high.
The set common requirement refers to a preset standard or threshold value for judging whether the similarity between the target data and the associated data is enough or not when the defect detection of the concrete structure is performed. Only data pairs meeting this criterion will be considered similar for subsequent defect analysis. For example, a system administrator may set a commonality index threshold, such as 0.8 (range 0-1), that is considered sufficiently similar for further defect detection only when the commonality index of the target data and the associated data reaches or exceeds 0.8.
The structural defect matching feature set is a data set containing various known concrete structural defects and their corresponding features. These features may be geometric anomalies, material property anomalies, construction joints, cracks, etc., which are associated with a particular defect type. For example, the structural defect matching feature set may include common defects of a series of concrete beams and features thereof, such as the features of a curved crack may include the location, length, width, shape, etc. of the crack. When the system detects a match with these features, the corresponding defect type can be identified.
The set defect labeling data refers to a set of known defects and labeling information thereof used for training and verifying a defect detection algorithm in a concrete structure defect detection system. Such data typically comes from expert human inspection and analysis results. For example, an expert team may perform detailed inspections on a set of concrete samples and record information about the type, location, size, etc. of defects present in each sample. These data are then used as set defect labeling data for training and verifying the defect detection capabilities of the system.
Defect state matching features refer to features or attributes in a set of structural defect matching features that are used to describe and identify a particular defect state. These features may include the shape, size, location, color, texture, etc. of the defect. For example, for cellular defects on a concrete surface, the defect state matching characteristics may include the size, depth, distribution density, surface texture, etc. of the honeycomb. The system identifies whether similar defect conditions exist by comparing the target data to these features.
The target structural defect detection result refers to a conclusion or report about structural defects, which is obtained by the concrete structural defect detection system after the target concrete BIM structural data is detected and analyzed. This result typically includes information about the type, location, severity, and possibly repair advice of the defect. For example, after the system detects the concrete columns of a building, a target structure defect detection result report may be generated, where the information indicating that a specific position of the column has a crack defect, the length and width of the crack, and the possible influence range are all included in the report.
The relevant nouns in step 150 are explained as follows.
The defect discrimination network is a deep learning network model specifically designed to identify and classify structural defects. It typically learns the characteristics of various defects by training a large amount of annotation data and is able to automatically detect and classify defects on new data based on these characteristics. For example, the defect discrimination network may be a Convolutional Neural Network (CNN) that receives as input an image of the concrete structure or BIM data and outputs the type and location of the defect. This network may be trained on millions of images with defect labels, so that similar defects can be accurately identified on new concrete structures.
The structural defect discriminating operation refers to a process of processing and analyzing the target concrete BIM structural data using a defect discriminating network to identify and classify structural defects that may exist therein. This process typically includes the steps of data preprocessing, feature extraction, defect detection, and classification. For example, in performing the structural defect discriminating operation, the system first performs necessary preprocessing, such as denoising, normalization, etc., on the target concrete BIM data. It then inputs the processed data into a defect discrimination network, which automatically extracts features associated with the defects and detects and classifies the defects based on those features. Finally, the system outputs a structural defect detection result containing defect type and position information.
The structural defect detection result refers to detailed information and conclusions about defects possibly existing in the BIM structural data of the target concrete, which are obtained after the processing and analysis of the defect discrimination network. This result typically includes the type, location, size, severity, and possibly repair advice of the defect. For example, the structural defect detection result may be a detailed report listing all defects detected on the target concrete structure. For each defect, the report provides its type (e.g., crack, honeycomb defect, etc.), location (e.g., specific components and coordinates), size (e.g., length, width, etc.), and severity (e.g., mild, moderate, severe, etc.). In addition, the report may also contain some repair suggestions to help engineers or architects formulate effective repair solutions.
In order to further understand the above technical solution, the following description is presented by a complete application scenario.
On a busy construction site, it becomes important to ensure the safety and quality of the construction structure as the progress of the construction progresses. In particular concrete structures, which carry the weight and various stresses of the entire building, it is therefore necessary to perform detailed and microscopic defect detection regularly. However, the traditional detection method, such as manual inspection and expert evaluation, is not only low in efficiency, but also easy to make mistakes, and cannot meet the requirements of modern building projects on speed and precision.
To solve this problem, project team decision introduces an intelligent detection system based on BIM (building information model) and deep learning. The system is capable of automatically receiving, processing and analyzing large amounts of BIM data to accurately identify potential defects in concrete structures.
When the system is started, firstly, the system receives the currently-incoming target concrete BIM structure data through a detection interface of the target task server. These data are detailed three-dimensional model information about the concrete structure to be inspected, including the geometry, dimensions, location, and material properties of the members such as beams, plates, columns, etc. These data are the basis for subsequent analysis and judgment by the system.
Next, the system performs attribute mining on the received BIM structure data. By the mining algorithm, the system is able to extract key attributes related to the target structure, such as component type, size, material strength, etc. These attributes are encoded into a target BIM structure attribute vector, providing basis for subsequent similarity comparisons and defect detection.
Then, the system searches the cloud BIM structure information set for similar initial concrete BIM structure data using the target BIM structure attribute vector. The cloud BIM structure information set is a huge database, and BIM data of various known concrete structures are stored. By comparing the similarity of the attribute vectors, the system is able to determine the associated concrete BIM structure data that is most similar to the target concrete BIM structure data.
After the associated data is found, the system can judge the commonality index. If the commonality index between the associated concrete BIM structure data and the target concrete BIM structure data meets the set commonality requirement (e.g. the similarity is above a certain threshold), the system considers them sufficiently similar that the defect state matching feature can be extracted from the set of structural defect matching features. The structural defect matching feature set is a data set containing various known concrete structural defects and corresponding features thereof, and the features are the basis for identifying and classifying the defects by the system.
Based on the extracted defect state matching features and associated concrete BIM structure data, the system can determine a target structure defect detection result. This result details the type, location, size and severity of all defects detected on the target concrete structure. The system then issues the result to the target task server for subsequent review and processing by the project team.
However, if the commonality index does not meet the set requirement, i.e., the similarity between the associated concrete BIM structure data and the target concrete BIM structure data is not high enough, the system cannot perform defect detection by direct comparison. In this case, the system would turn to performing structural defect discrimination operations on the target concrete BIM structural data using a pre-trained defect discrimination network. The defect discrimination network is a deep learning model that is capable of automatically extracting features and detecting potential defects by learning a large amount of annotation data.
Through the processing and analysis of the defect discrimination network, the system can generate the structural defect detection result of the target concrete BIM structural data. This result also details the information of all defects detected and sent to the target task server for review and processing by the project team.
Through the application of the technical scheme, the intelligent detection system for the defects of the concrete structure can efficiently process and analyze a large amount of BIM data and accurately identify potential defects in the concrete structure. The method not only greatly improves the detection efficiency and accuracy, but also provides powerful support for subsequent repair and maintenance work. Meanwhile, the automation and intelligent characteristics of the system also lighten the workload of project team and improve the overall quality and safety of the building project.
In an alternative embodiment, the initial concrete BIM structure data includes a BIM element unit and a concrete space structure diagram, and the set defect labeling data includes element trend characteristics and mechanical conduction data; the step 210-step 250 is further included before the step of obtaining the structural defect matching feature set.
Step 210, constructing a first defect state matching feature between the BIM element unit and the element trend feature thereof.
Step 220, determining mechanical conduction data for the BIM element unit, and constructing a second defect state matching feature between the BIM element unit and the mechanical conduction data.
And 230, constructing a third defect state matching characteristic between the concrete space structure diagram and mechanical conduction data corresponding to the concrete space structure diagram.
And 240, constructing a fourth defect state matching feature between the concrete space structure diagram and the corresponding element trend feature.
Step 250, determining a structural defect matching feature set based on the first defect state matching feature, the second defect state matching feature, the third defect state matching feature, and the fourth defect state matching feature.
In an alternative embodiment, the intelligent detection system for the defects of the concrete structure adopts a specific technical scheme to implement the detection flow. In this scheme, the initial concrete BIM structure data is defined in detail to comprise BIM element units and a concrete space structure diagram, and the set defect marking data covers element trend characteristics and mechanical conduction data. To construct an accurate and comprehensive structural defect matching feature set, the system performs the detailed operations of steps 210 through 250 prior to acquiring the feature set.
In step 210, the system first builds a first defect state matching feature between the BIM element unit and its element trend feature. Here, BIM element units refer to basic elements constituting a concrete structure, such as beams, plates, columns, etc., and element trend features describe the trend or behavior pattern of these elements in the structure. By further analyzing the BIM data and corresponding trend features, the system can identify patterns associated with particular defects, thereby constructing first defect state matching features.
Next, in step 220, the system determines mechanical conduction data for the BIM element and constructs a second defect state matching feature between the BIM element and the mechanical conduction data. The mechanical conductivity data describes the mechanical response of the structure when subjected to external forces, such as stress distribution, deformation, etc. By correlating these data with the BIM element units, the system is able to further understand the behavior of the structure under different stress conditions and identify features associated with mechanical defects.
In step 230, the system builds a third defect state matching feature between the concrete space block diagram and the corresponding mechanical conductivity data. The concrete space block diagram provides a three-dimensional view of the structure, including the spatial location and interrelationship of the components. By combining the mechanical conduction data with the spatial structure diagram, the system can analyze the mechanical properties and defect modes of the structure on the whole layer.
Finally, in step 240, the system builds a fourth defect state matching feature between the concrete space block diagram and its corresponding element trend feature. This step correlates the spatial structure map with elemental trend features, enabling the system to identify and analyze structural defects at a more macroscopic level.
After the four above steps are completed, the system determines a structural defect matching feature set based on the first defect state matching feature, the second defect state matching feature, the third defect state matching feature, and the fourth defect state matching feature in step 250. The feature set is a comprehensive data set containing feature information related to structural defects from multiple angles and layer analyses.
After the structural defect matching feature set is obtained, the system can execute subsequent common index judgment and defect detection operations according to the need. By comparing the commonality index between the target concrete BIM structure data and the associated concrete BIM structure data, the system is able to determine whether to use a direct comparison method for defect detection or to switch to more complex analysis using a defect discrimination network. Finally, the system will generate detailed structural defect detection results and issue to the target task server for further processing and application.
In some preferred embodiments, the determining, according to the target BIM structure attribute vector described in step 130, associated concrete BIM structure data corresponding to the target concrete BIM structure data from initial concrete BIM structure data of a cloud BIM structure information set includes steps 131 to 132.
Step 131, identifying a building scene tag to which the target concrete BIM structure data belongs; and determining initial concrete BIM structure data under the building scene label in the cloud BIM structure information set.
And 132, determining associated concrete BIM structure data corresponding to the target concrete BIM structure data from the initial concrete BIM structure data under the building scene label according to the target BIM structure attribute vector.
In some preferred embodiments, the intelligent concrete structure defect detection system performs step 130, further refined to step 131 and step 132, to ensure that the associated data corresponding to the target concrete BIM structure data is more accurately determined from the cloud BIM structure information set.
In step 131, the system first identifies the building scene tag to which the target concrete BIM structure data belongs. Building scene tags are descriptions of features of building type, use, or environment, such as residential buildings, office buildings, bridges, and the like. By identifying the tag to which the target data belongs, the system can reduce the search range and improve the data processing efficiency. After the tag is determined, the system can further screen out initial concrete BIM structure data which belongs to the building scene tag in the cloud BIM structure information set. This one step helps to ensure that the data of the subsequent comparison is performed in the same context of the building scene, thereby improving the accuracy and relevance of the comparison.
Next, in step 132, the system determines associated concrete BIM structure data corresponding to the target concrete BIM structure data from the initial concrete BIM structure data under the building scene label that has been screened based on the target BIM structure attribute vector. The correspondence is determined based on the similarity of the BIM structure attribute vectors. The system compares the similarity between the target vector and each vector in the cloud data set, and selects BIM structure data corresponding to one or more vectors with highest similarity as associated data. The associated data have higher similarity with the target data in terms of structure, material, size and the like, so that the associated data can be used for subsequent commonality index judgment and defect detection operation.
Through the execution of the two steps, the system can more accurately determine the associated data corresponding to the target concrete BIM structure data from the cloud BIM structure information set, and a reliable data basis is provided for subsequent structural defect detection. Meanwhile, the data screening and matching method based on the building scene tag and the BIM structure attribute vector improves the automation and intelligent level of the system and reduces the requirements of manual intervention and judgment.
In still other exemplary embodiments, the determining, based on the structural defect matching feature set and the associated concrete BIM structural data, the target structural defect detection result corresponding to the associated concrete BIM structural data described in step 140 includes steps 141-142.
And 141, determining target set defect labeling data corresponding to the associated concrete BIM structure data from the structure defect matching feature set.
And 142, when the target set defect labeling data comprises mechanical conduction data aiming at the associated concrete BIM structure data, processing the target concrete BIM structure data according to the mechanical conduction data to obtain a target structure defect detection result.
In still other exemplary embodiments, the intelligent concrete structure defect detection system may be subdivided into steps 141 and 142 when executing step 140 to determine a target structure defect detection result corresponding to the associated concrete BIM structure data. The two steps ensure that the system can accurately identify and process the defect marking information related to the specific associated data, and accordingly obtain an accurate structural defect detection result.
In step 141, the system determines target set defect labeling data corresponding to the associated concrete BIM structural data from the previously constructed set of structural defect matching features. The structural defect matching feature set is a comprehensive data set containing a plurality of defect features and corresponding labeling information. The system can find the corresponding set defect marking data by comparing the characteristics of the associated data with the information in the data set. These data provide critical information regarding the type, location, severity, etc., of defects that may be present in the associated concrete BIM structure data.
Next, in step 142, the system checks if the target set defect labeling data includes mechanical conductivity data for the associated concrete BIM structure data. Mechanical conduction data is an important parameter describing the response and performance of a structure in a stressed state, and is important for accurately evaluating the integrity and safety of the structure. If such data is indeed included in the target set defect labeling data, the system further processes and analyzes the target concrete BIM structure data according to the mechanical conduction data.
During processing, the system may employ various algorithms and models to parse the mechanical conduction data to extract critical information related to structural defects. For example, the system may analyze stress distribution, deformation patterns, or vibration characteristics, etc. in the data to identify potential cracks, deformations, instabilities, or other types of structural defects. Finally, the system integrates the analysis results into target structural defect detection results, and details information such as types, positions, sizes, severity and the like of all defects detected in the target concrete BIM structural data.
Through the execution of the steps, the intelligent detection system for the defects of the concrete structure can accurately identify and evaluate potential defects in the structure by fully utilizing the information in the structural defect matching feature set and the associated concrete BIM structure data. This not only improves the level of automation and intelligence of the detection process, but also provides powerful data support for subsequent repair and maintenance work.
In yet other alternative embodiments, the step 110 of obtaining the target concrete BIM structure data currently entered by the detection interface of the target task server in the concrete structure defect detection system includes steps 111-112.
And 111, generating at least one template concrete BIM structure data according to the building scene label of the target task server, and transmitting the template concrete BIM structure data to the target task server in a detection interface in the concrete structure defect detection system.
And 112, selecting and processing the transmitted template concrete BIM structure data in the detection interface according to the target task server, and acquiring the target concrete BIM structure data currently transmitted by the target task server in the detection interface.
In other alternative embodiments, the intelligent concrete structure defect detection system may take a form based on template data interaction when executing step 110 to obtain the target concrete BIM structure data currently transmitted by the detection interface of the target task server in the concrete structure defect detection system. This process is specifically subdivided into step 111 and step 112.
In step 111, the system will first generate at least one template concrete BIM structure data based on the building scene tags of the target task server. Building scene tags are identifications describing the type, purpose, or specific environmental characteristics of a building, such as "high-rise residential building", "commercial complex", or "subway station", etc. These tags assist the system in understanding the type and structural features of the building that are of interest to the target task server. Based on the tags, the system extracts typical or standard concrete BIM structure data corresponding to the tags from a BIM database or cloud BIM resource library built in the system as template data. These template data are pre-designed, representative and generic BIM models that reflect the basic structure and features of a certain class of building.
After the template data are generated, the system transmits the template concrete BIM structure data to a target task server in a detection interface of the concrete structure defect detection system. The aim of the step is to enable the target task server to select one or more most suitable target concrete BIM structure data from the provided template data according to the requirements and actual conditions of the target task server to carry out subsequent defect detection.
Next, in step 112, the system waits and listens to the result of the picking process of the transmitted concrete BIM structure data of the template in the detection interface by the target task server. The target task server evaluates the applicability and relevance of each template data according to its own algorithm, rules or user input, and selects one or more data most meeting the current detection task requirements as target concrete BIM structure data. Once selected, the data is transmitted back to the intelligent detection system for the defects of the concrete structure by the target task server through the detection interface.
After the system receives the data, the data are regarded as the current incoming target concrete BIM structure data, and the subsequent defect detection operation is carried out according to the data. By the method, the system can perform effective data interaction and cooperation with the target task server, and the acquired BIM structure data is ensured to be closely related to the current detection task and to be accurate and reliable.
In other preferred embodiments, the acquiring the target task server further includes steps 310-330 after the target concrete BIM structure data currently entered by the detection interface in the concrete structure defect detection system.
And 310, carrying out concrete BIM structure data quality analysis on the target concrete BIM structure data.
And 320, transmitting concrete BIM structure data under a template BIM element to the target task server when the quality analysis report accords with a set index, wherein the template BIM element is different from BIM elements contained in the target concrete BIM structure data.
And 330, when the quality analysis report does not accord with the set index, transmitting basic concrete BIM structure data under a target BIM element to the target task server, wherein the target BIM element is associated with BIM elements contained in the target concrete BIM structure data.
In other preferred embodiments, the intelligent concrete structure defect detection system further performs steps 310 to 330 after obtaining the target concrete BIM structure data currently transmitted by the detection interface of the target task server in the concrete structure defect detection system, so as to ensure the accuracy and effectiveness of the data, and provide corresponding BIM structure information according to the data quality.
In step 310, the system performs mass analysis on the obtained target concrete BIM structure data. This process involves the evaluation of various aspects of the integrity, consistency, accuracy and readability of the BIM data. The system uses a preset quality assessment algorithm and rules to check and analyze geometric information, attribute information, association relations and the like in the data so as to generate a quality analysis report. The report may detail various problems in the data, such as missing components, incorrect attribute assignments, inconsistent associations, etc., and give corresponding quality scores.
Next, in step 320, the system determines whether the quality resolution report meets the set criteria. The set index is a set of criteria preset by the system and used for measuring whether the quality of BIM data meets the requirement of subsequent processing. If all indexes in the quality analysis report meet the set indexes, the quality of the target concrete BIM structure data is higher, and the method can be directly used for subsequent defect detection operation. At this time, the system transmits concrete BIM structure data under the template BIM element to the target task server. The template BIM elements herein are another set of BIM elements that are different from the BIM elements contained in the target concrete BIM structure data, which may represent a broader or more specific building structure and feature. The purpose of transmitting these data is to provide additional reference information to help the target task server more fully understand and analyze the target concrete BIM structure data.
However, if the quality analysis report does not meet the set index, it is indicated that the target concrete BIM structure data has quality problems in step 330, and may not be directly used for the subsequent defect detection operation. At this point, the system will take remedial action to transmit the underlying concrete BIM structure data under the target BIM element to the target mission server. The target BIM elements are a set of BIM elements associated with the BIM elements contained in the target concrete BIM structure data, which may represent the same or similar building structure and characteristics. The purpose of transmitting such data is to provide a basic or alternative BIM model for use by the target task server when the original data cannot be processed. By the method, the system can reduce the influence of the data quality problem on subsequent processing to a certain extent, and ensure the smooth proceeding of the defect detection operation.
In some alternative embodiments, the performing BIM structure attribute mining on the target concrete BIM structure data described in step 120 to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data includes: and performing BIM structure attribute mining on the target concrete BIM structure data through a BIM structure attribute mining algorithm to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data.
In some alternative embodiments, the concrete structure defect intelligent detection system, when executing step 120, processes the target concrete BIM structure data using a specific BIM structure attribute mining algorithm. The purpose of this step is to extract key structural attribute information from the complex BIM data and convert it into a mathematical vector form that can be used for subsequent analysis and comparison.
Specifically, the system will first invoke the BIM structure attribute mining algorithm. This is an algorithm specifically designed for processing BIM data that is capable of identifying and extracting various structural attributes in the BIM model, such as geometry, dimensions, materials, connection patterns, etc. of the components. These attributes are key factors in describing building structural features and performance and are critical to subsequent defect detection.
The system then provides the target concrete BIM structure data as input to the BIM structure attribute mining algorithm for processing. The algorithm analyzes each component in the data one by one according to preset rules and logic, and extracts the corresponding structural attribute information. This process may involve traversing and parsing the various levels and components in the BIM model to ensure that all relevant structural attributes are accurately identified and extracted.
Finally, the system converts the extracted structural attribute information into a mathematical vector form to generate a target BIM structural attribute vector. The vector is a multi-dimensional array of values, each dimension representing a particular structural attribute, and the values reflect the concrete or quantized values of that attribute in the target concrete BIM structural data. In this way, the system is able to structure complex BIM data into a set of comparable and computable numerical vectors, providing powerful data support for subsequent defect detection and analysis.
Under some optional design ideas, the method further comprises steps 410-460 before performing BIM structure attribute mining on the target concrete BIM structure data by using a BIM structure attribute mining algorithm to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data.
Step 410, an algorithm tuning sample is obtained, wherein the algorithm tuning sample comprises a plurality of concrete BIM structure data examples.
Step 420, for each concrete BIM structure data instance, generates a set of positive example tuples of the concrete BIM structure data instance, the set of positive example tuples comprising the concrete BIM structure data instance and at least one positive training data having a common characteristic with the concrete BIM structure data instance.
Step 430, sorting the concrete BIM structure data examples and each positive training data respectively to construct at least one first training example binary set of the concrete BIM structure data examples; and constructing at least one second training example binary group of the concrete BIM structure data example according to the concrete BIM structure data example and the data in the positive example binary group set of each concrete BIM structure data example.
Step 440, performing BIM structure attribute mining on the data in the first training example binary group and the second training example binary group through a preset BIM structure attribute mining algorithm, so as to obtain BIM structure attribute vectors of the data in the first training example binary group and the second training example binary group.
Step 450, determining attribute vector differences of the first training example binary group based on BIM structure attribute vectors of each content in the first training example binary group; and determining the attribute vector difference of the second training example binary group based on the BIM structure attribute vector of the content in the second training example binary group.
Step 460, optimizing algorithm variables of a preset BIM structure attribute mining algorithm according to the attribute vector difference of the first training example binary group and the attribute vector difference of the second training example binary group, and obtaining an optimized BIM structure attribute mining algorithm.
Under some optional design ideas, the intelligent detection system for the defects of the concrete structure can be subjected to an algorithm debugging and optimizing process before the BIM structure attribute mining is executed. This process aims to verify and adjust the performance of the BIM structure attribute mining algorithm through the sample data to ensure that it can accurately extract critical structure attribute information from the concrete BIM structure data.
First, in step 410, the system obtains algorithm tuning examples including a plurality of concrete BIM structure data examples. These example data are extracted from the actual BIM model, are representative and diverse, and can cover different types of building structures of varying complexity.
Next, in step 420, the system generates a set of positive example tuples for each concrete BIM structure data example. This set consists of two parts: firstly, concrete BIM structure data examples per se, and secondly, at least one positive training data with common characteristics with the examples. Commonality may refer to the same structural type, similar geometry, the same material properties, and the like. Positive training data is data that is similar to, but not exactly the same as, example data for providing positive reference information during training.
Then, in step 430, the system sorts concrete BIM structure data examples with each positive training data to construct at least one first training example doublet. Meanwhile, the system also constructs at least one second training example binary group according to the data in the positive example binary group set of the concrete BIM structure data example and other residual examples. These training example tuples will be used for subsequent algorithm training and verification.
In step 440, the system performs BIM structure attribute mining on the data in the first training example tuple and the second training example tuple respectively using a preset BIM structure attribute mining algorithm. This process extracts the key structure attribute information for each data and converts it into a mathematical vector form, i.e., a BIM structure attribute vector.
Next, in step 450, the system calculates the difference between the attribute vectors of the two sets of data based on the BIM structure attribute vectors of the contents of the first training example set and the second training example set. The attribute vector difference can reflect the similarity and the difference of the structural attributes among different data, and is an important index for evaluating the performance of the algorithm.
Finally, in step 460, the system optimizes algorithm variables of the preset BIM structure attribute mining algorithm according to the attribute vector differences of the first training example binary set and the attribute vector differences of the second training example binary set. The optimization aims at minimizing attribute vector difference and improving the processing capacity and accuracy of the algorithm on concrete BIM structure data. The optimized BIM structure attribute mining algorithm is used for subsequent defect detection and analysis tasks.
In more detail, for step 420, the process of generating its set of positive example tuples is a key step in algorithm debugging and optimization for each concrete BIM structure data example. The construction of this set aims to enhance the recognition and processing power of the algorithm on similar but not exactly identical BIM structures by introducing positive training data with common features to the original examples.
(1) Concrete BIM structure data examples were selected: first, a concrete BIM structure data is selected from a debug sample library as an example of the current process. This example may be a BIM model of a beam, plate, column or any other concrete member.
(2) Analysis of example features: and (3) carrying out detailed analysis on the selected BIM structure data example, and extracting key characteristics such as geometric shapes, sizes, material properties, spatial relations and the like. These features constitute the structural attributes of the BIM model, which are critical to subsequent algorithmic training.
(3) Searching positive training data: next, the system searches the sample library or other available data sources for other BIM data that has common characteristics with the current BIM structure data example. These data are referred to as "positive training data" because they are similar to, but not exactly the same as, the original examples in some key features. This similarity makes positive training data an effective reference for algorithm learning. The determination of the commonality characteristics may be based on a variety of criteria, such as similarity of geometric shapes, degree of matching of material properties, consistency of spatial relationships, and the like. The system will evaluate and select the most appropriate positive training data according to preset criteria.
(4) Constructing an active example binary group: once at least one positive training data is found that has common characteristics with the current BIM structure data instance, the system pairs them together to form a "positive instance tuple". This tuple contains the original BIM structure data instance and a similar positive training data.
(5) Forming a positive example set of tuples: by repeating the above steps, the system generates one or more positive example doublets for each concrete BIM structure data example. These tuples are grouped together to form a positive example tuple set of the current BIM structure data example. This set of positive example tuples plays an important role in subsequent algorithm training and evaluation. They provide a challenging training scenario for algorithms that help the algorithm learn how to distinguish and process BIM structure data that has similar characteristics but is not exactly the same. In this way, the algorithm can gradually optimize its ability to process complex BIM data, improving the accuracy and reliability of defect detection.
In more detail, for step 430, constructing training example doublets is a core step in algorithmic debugging of concrete BIM structure data examples. This process involves pairing BIM structure data examples with positive training data to form a data set for algorithm training. In the following, i will explain in detail how the first training example binary set and the second training example binary set are constructed and give specific examples.
(1) Construction of a first training example tuple
Firstly, starting from concrete BIM structure data examples, for each example, positive training data is selected from a corresponding positive example binary group set to pair. This pairing is the process of building the first training example doublet.
There is a concrete BIM structure data example a, which represents a particular beam member. In the set of positive example tuples, positive training data B is found that has common characteristics (e.g., similar geometry and material properties) to example a. Thus, example A is paired with positive training data B, forming a first training example doublet (A, B).
This tuple will be used to train the algorithm, helping the algorithm learn how to identify and process BIM structure data similar to example a. Through multiple such dyadic training, the algorithm can gradually grasp the ability to identify common features.
(2) Construction of a second training example tuple
In addition to pairing with positive training data to form a first training example doublet, a second training example doublet is constructed using the remaining concrete BIM structure data examples. These tuples will be used to further train the algorithm, improving its ability to process different BIM structure data.
Continuing with example a above, there are other concrete BIM structure data examples C, D, E, etc., in addition to the positive training data B. These examples may differ from example a in some features but still have some similarity. To enhance the generalization ability of the algorithm, example a would be paired with example C, D, E, etc., respectively, to form a plurality of second training example doublets, such as (a, C), (a, D), (a, E), etc.
These second training example tuples will provide more training scenarios for the algorithm, helping the algorithm learn how to handle BIM structure data that is slightly different from example a but still relevant. Through such training, the algorithm may gradually increase its ability to process complex and variable BIM data.
In summary, by constructing the first training example binary set and the second training example binary set, a comprehensive and rich training data set can be provided for the algorithm, so that the processing capability and accuracy of the algorithm on concrete BIM structure data are improved.
In more detail, optimizing the BIM structure attribute mining algorithm is a key step to improve detection accuracy and efficiency in the concrete structure defect intelligent detection system for step 460. This process relies on the analysis and exploitation of the difference in attribute vectors of the first training example doublet and the second training example doublet. How to optimize the algorithm variables based on these differences to obtain more accurate BIM structure attribute mining capabilities will be described in detail below.
First, the concept of explicit attribute vector differences is required. In a previous step, the system has extracted a BIM structure attribute vector for each training example tuple (including the first training example and the second training example) by a BIM structure attribute mining algorithm. These vectors are a numerical representation of the structural features of the BIM model, reflecting the properties of the model in terms of geometry, materials, spatial relationships, etc. Attribute vector differences refer to the numerical differences between different BIM structure data instances over these attribute vectors.
Next, the system calculates a difference in attribute vectors for the first training example doublet and the second training example doublet. These differences quantify similarities and differences between BIM structure data examples, providing an important basis for algorithm optimization. By analyzing these differences in comparison, it can be found that the algorithm may have deficiencies or deviations in processing certain specific types of BIM structure data.
Then, based on the calculated attribute vector differences, the system adjusts the algorithm variables of the BIM structure attribute mining algorithm. These algorithm variables typically include weights, thresholds, parameters, etc., that control the behavior and decision process of the algorithm in processing the BIM structure data. The goal of the optimization is to enable the algorithm to exhibit good performance and accuracy in processing various types of BIM structure data.
In particular, if the system finds that the algorithm has difficulty in processing certain BIM structure data with certain common characteristics (e.g., misjudgments, missed judgments, etc.), it may increase the weight of those characteristics in the algorithm or adjust the relevant thresholds and parameters to make the algorithm more sensitive and accurate to those characteristics. Conversely, if the algorithm is overfitted on some unimportant features, the system may decrease the weight of those features or adjust the relevant parameters to reduce the dependence of the algorithm on those features.
Finally, by continuously iterating and adjusting the algorithm variables, the system can gradually optimize the performance and accuracy of the BIM structure attribute mining algorithm. This process may involve multiple trials and verifications to ensure that the optimized algorithm is capable of performing well in a variety of practical application scenarios.
In summary, optimizing the BIM structure attribute mining algorithm based on the difference in the attribute vectors of the first training example doublet and the second training example doublet is a complex but critical process. The method is beneficial to improving the processing capacity and accuracy of the algorithm on concrete BIM structure data, and provides a more reliable basis for subsequent defect detection and analysis.
In some examples, the concrete BIM structure data examples include BIM element units and concrete space structure diagram examples; the generating the positive example binary set of concrete BIM structure data examples described in step 420 includes: respectively carrying out image data derivatization on the BIM element unit and the concrete space structure diagram example to obtain a derivatization element unit of the BIM element unit and a concrete space structure derivatization diagram of the concrete space structure diagram example; determining a background image in the concrete space structure diagram example based on the set element word vector and a preset BIM element unit; updating the background image in the concrete space structure diagram example to obtain an updated concrete space structure diagram example; acquiring quantization characteristics of the BIM element unit; determining positive training data of the BIM element unit according to the derivative element unit and the quantization characteristic of the BIM element unit; and determining positive training data of the concrete space structure diagram example according to the concrete space structure derivative diagram of the concrete space structure diagram example and the updated concrete space structure diagram example.
In some specific application examples, when concrete BIM (Building Information Modeling, building information model) structural data is processed, these data examples include not only BIM element units, but also concrete space structure diagram examples. To generate a positive example set of tuples of this data, the system needs to perform a complex series of operations. The following is a detailed illustration of these operations.
First, the system will derive image data for the BIM element unit and concrete space structure diagram examples, respectively. This process involves converting or enhancing the original image data to extract more useful information or features. For BIM element units, the system generates its derivative element units, which may include extracting key information about its geometry, size, location, etc., and representing this information in the form of values or vectors. For the concrete space structure diagram example, the system may generate its concrete space structure derivative map, which may involve filtering, edge detection, feature extraction, etc. the image to highlight key structural features in the image.
Next, the system determines a background map in the concrete space structure diagram example from the preset element word vector and BIM element unit. An element word vector is a set of predefined vectors that represent key features or attributes in a BIM element unit. The system will match the various parts in the concrete space block diagram example with these element word vectors to determine which parts belong to the background map. These background maps are typically parts that do not contain critical structural information, such as sky, ground, etc. The system then updates the background images, for example, by replacing them with standard background images or performing image enhancement processing to obtain an updated concrete space structure diagram example.
After determining the derived data of the BIM element unit and the concrete space structure diagram example, the system further obtains the quantization characteristic of the BIM element unit. These quantitative features are key attributes or features that represent BIM element elements in numerical form, such as length, width, height, volume, weight, etc. These features will be used for subsequent algorithm training and evaluation.
Next, the system determines its positive training data based on the derived element units and the quantized features of the BIM element units. Positive training data refers to other BIM data that have common characteristics with the original BIM element, and that are similar but not identical to the original data in some key characteristics. The system will search for and select these positive training data by comparing the similarity of the derived element units and the quantized features. For example, the system may select other BIM element elements that have similar geometry, size and location as the original BIM element as the positive training data.
Finally, the system may determine its positive training data based on the concrete space structure derivative graph of the concrete space structure graph example and the updated concrete space structure graph example. Similar to the BIM element units, the system will search and select other concrete space structure diagrams with common features as positive training data by comparing the similarity of derivative and updated graphs. These positive training data will be used in subsequent algorithm training and evaluation to improve the recognition and processing power of the algorithm for similar but not identical concrete space structure diagrams.
In summary, through this complex series of operations and processing steps, the system is able to generate a positive example binary set of concrete BIM structure data examples. These sets will provide a rich and diverse data resource for subsequent algorithm training and evaluation, helping to improve the processing power and accuracy of the algorithm to concrete BIM structure data.
In other possible embodiments, the optimizing the algorithm variable of the preset BIM structure attribute mining algorithm according to the attribute vector difference of the first training example binary set and the attribute vector difference of the second training example binary set described in step 460 to obtain an optimized BIM structure attribute mining algorithm includes steps 461-464.
Step 461, integrating the attribute vector differences of the second training example binary groups to obtain integrated second attribute vector differences.
Step 462, for each first training example binary set, determining local training cost information of the first training example binary set according to the integrated second attribute vector difference and the attribute vector difference of the first training example binary set.
And 463, integrating the local training cost information of each first training example binary group to obtain global training cost information.
And 464, optimizing algorithm variables of a preset BIM structure attribute mining algorithm according to the global training cost information to obtain an optimized BIM structure attribute mining algorithm.
In other possible embodiments, step 460 details an optimization process that exploits the difference in attribute vectors of the first training example doublet and the second training example doublet to refine the preset BIM structure attribute mining algorithm. The following is a detailed explanation of steps 461 to 464 so that the implementation details of this solution can be more clearly understood.
In step 461, the system processes the second training example tuple. These tuples typically contain the original BIM structure data instance and its corresponding variant version. The system calculates the attribute vector differences between the two data examples in each second training example tuple. The attribute vector differences reflect the degree of change in the key attributes of the BIM structure data. The system then integrates the difference information to form a comprehensive second attribute vector difference. The process of integration may include calculating an average, maximum, minimum, or other statistic of the differences to obtain a composite index that is representative of all second training example bigram differences.
Next, in step 462, the system focuses on the first training example doublet. These tuples are typically composed of original BIM structure data examples and their positive training data. For each first training example tuple, the system determines a local training cost information in combination with the previously calculated integrated second attribute vector differences and the tuple's own attribute vector differences. Here, the local training cost information may be understood as a loss or error that the algorithm generates when processing this particular tuple. It may be derived by calculating the norm of the difference vector, distance metric, or other loss function. This local cost information reflects challenges that the algorithm may encounter when processing similar data.
Then, in step 463, the system may aggregate the local training cost information of all the first training example tuples to form a global training cost information. This process may include a weighted average, summation, or other aggregation operation of all local cost information. The global training cost information provides a comprehensive view that demonstrates the overall performance of the algorithm when processing the entire training data set.
Finally, in step 464, the system optimizes a pre-set BIM structure attribute mining algorithm using the global training cost information. The optimization process may involve adjusting parameters, weights, model structures, or other configurations of the algorithm to minimize global training cost information. This may include iteratively improving the performance of the algorithm using gradient descent, random gradient descent, genetic algorithm, or other optimization techniques. The optimized BIM structure attribute mining algorithm can be better adapted to training data, and higher accuracy and efficiency are shown when similar data are processed.
In summary, through these detailed steps and explanations, it can be further understood how to use the attribute vector differences to optimize the technical solution of the BIM structure attribute mining algorithm. The optimization method is beneficial to improving the performance and accuracy of the algorithm when processing complex BIM structure data, and provides more reliable support for concrete structure defect detection and analysis.
In some independent embodiments, after performing the structural defect discriminating operation on the target concrete BIM structural data through the defect discriminating network described in step 150 to obtain the structural defect detection result corresponding to the target concrete BIM structural data, the method further includes step 160.
And 160, generating a concrete structure maintenance scheme according to the structural defect detection result.
In some independent embodiments, when the system performs the structural defect determining operation on the target concrete BIM structural data through the defect determining network and successfully obtains the structural defect detection result corresponding to the target concrete BIM structural data, the system further performs step 160.
In step 160, the system generates a concrete structure maintenance plan based on the previously obtained structural defect detection results. This maintenance scheme is designed to address or mitigate structural defects detected in the target concrete BIM structural data.
Specifically, the system may first analyze the structural defect detection results to determine the type, location, and severity of defects present in the target concrete structure. This information is the key basis for the development of maintenance schemes.
Then, the system selects corresponding maintenance measures from a preset maintenance strategy library according to the type and severity of the defects. For example, for crack-like defects, the system may suggest maintenance measures such as filling, sealing, or reinforcement; for corrosion-type defects, the system may recommend surface coating protection, replacement of corrosion components, or maintenance methods using corrosion-resistant materials, etc.
Meanwhile, the system can further optimize and adjust selected maintenance measures by considering the position of the defect and the overall performance requirement of the concrete structure. For example, for critical load bearing site defects, the system may choose a more conservative and safe maintenance strategy to ensure overall stability of the structure.
In addition, the system can customize and evaluate the feasibility of the maintenance scheme according to the actual demands and resource conditions of the user. This includes consideration of maintenance costs, time, labor, and materials, among other factors, ensuring that the resulting maintenance solution is both economical and practical.
Finally, the system generates a detailed concrete structure maintenance plan report including defect summaries, maintenance action recommendations, implementation steps, expected effects, and notes. The report can provide comprehensive maintenance guidance for users, help them to effectively solve the defect problem in the concrete structure in time and ensure the safety and stability of the building.
In summary, the system automatically generates a concrete structure maintenance scheme according to the previously obtained structural defect detection results to solve or alleviate the structural defects identified in the target concrete BIM structure data.
Based on this, in other independent embodiments, the method further comprises: based on the structural defect detection result, determining the type, the position and the severity of the defect in the target concrete structure, selecting initial maintenance measures matched with the type and the severity of the defect from a preset maintenance strategy library, and then optimally adjusting the initial maintenance measures by combining the defect position and the overall performance requirement of the target concrete structure; customizing and carrying out feasibility assessment processing on the initial maintenance measures according to user requirements and resource conditions, wherein the feasibility assessment processing comprises assessment of maintenance cost, time, manpower and material factors, so as to obtain target maintenance measures; and generating a concrete structure maintenance scheme report containing defect summary, maintenance measure suggestion, implementation steps, expected effect and notice according to the target maintenance measure.
The design is that the type, the position and the severity of the defect in the target concrete structure are determined based on the structural defect detection result, and initial maintenance measures matched with the defect type and the severity are selected from a preset maintenance strategy library, and the process has the following beneficial effects:
improving the pertinence of maintenance: through detailed analysis of structural defects, the selected maintenance measures can be matched with the actual defect types and severity, so that the pertinence and the effect of maintenance work are improved;
Optimizing resource configuration: the initial maintenance measures are optimized and adjusted by combining the defect positions and the overall performance requirement of the target concrete structure, so that the maintenance work can be ensured to be more reasonable in resource allocation while the structural safety is met, and unnecessary waste is avoided;
The actual requirements of users are met: the customization and feasibility evaluation processing are carried out on the initial maintenance measures according to the user requirements and the resource conditions, so that the final maintenance scheme can be ensured to meet the actual requirements of the user, and the practicability and the acceptability of the scheme are improved within the resource condition range of the user;
And the feasibility of a maintenance scheme is improved: the maintenance cost, time, manpower and material factors are evaluated to obtain target maintenance measures, so that the implementation of a maintenance scheme can be ensured to be feasible, and the risk of encountering unpredictable difficulties in the implementation process is reduced;
Providing comprehensive maintenance guidance: generating a concrete structure maintenance scheme report containing defect summary, maintenance measure suggestions, implementation steps, expected effects and notes according to the target maintenance measures, providing comprehensive maintenance guidance for users, helping the users to better understand and execute the maintenance scheme, and improving the efficiency and quality of maintenance work;
In summary, the process provides a scientific, reasonable and practical concrete structure maintenance scheme generating method by comprehensively considering the actual condition of the structural defect, the requirement and resource condition of the user and the feasibility of the maintenance scheme, which is beneficial to improving the safety and stability of the concrete structure and prolonging the service life of the concrete structure.
Fig. 2 shows a block diagram of a concrete structure defect detection system 300, comprising: memory 310 for storing program instructions and data; a processor 320, coupled to the memory 310, executes instructions in the memory 310 to implement the methods described above.
Further, a computer storage medium is provided containing instructions which, when executed on a processor, implement the above-described method.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. The apparatus and method embodiments described above are merely illustrative, for example, flow diagrams and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present invention may be integrated together to form a single part, or each module may exist alone, or two or more modules may be integrated to form a single part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a network device, or the like) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes. It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A concrete structure defect detection method in combination with BIM, characterized in that it is applied to a concrete structure defect detection system, the method comprising:
Acquiring target concrete BIM structure data currently transmitted by a detection interface of a target task server in the concrete structure defect detection system;
Performing BIM structure attribute mining on the target concrete BIM structure data to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data;
Determining associated concrete BIM structure data corresponding to the target concrete BIM structure data from initial concrete BIM structure data of a cloud BIM structure information set according to the target BIM structure attribute vector;
When the commonality index between the associated concrete BIM structure data and the target concrete BIM structure data accords with a set commonality requirement, a structure defect matching feature set is obtained, wherein the structure defect matching feature set comprises defect state matching features between initial concrete BIM structure data and set defect labeling data; determining a target structural defect detection result corresponding to the associated concrete BIM structure data based on the structural defect matching feature set and the associated concrete BIM structure data, and issuing the target structural defect detection result for the target concrete BIM structure data to the target task server;
When the commonality index between the associated concrete BIM structure data and the target concrete BIM structure data does not meet the set commonality requirement, carrying out structure defect judging operation on the target concrete BIM structure data through a defect judging network to obtain a structure defect detection result corresponding to the target concrete BIM structure data, and issuing the structure defect detection result for the target concrete BIM structure data to the target task server.
2. The method of claim 1, wherein the initial concrete BIM structure data includes BIM element units and a concrete space structure map, and the set defect labeling data includes element trend characteristics and mechanical conduction data; before the obtaining the structural defect matching feature set, the method further comprises:
Constructing a first defect state matching feature between the BIM element unit and the element trend feature thereof;
Determining mechanical conduction data for a BIM element unit, and constructing a second defect state matching feature between the BIM element unit and the mechanical conduction data;
constructing a third defect state matching characteristic between a concrete space structure diagram and mechanical conduction data corresponding to the concrete space structure diagram;
Constructing a fourth defect state matching feature between the concrete space structure diagram and the corresponding element trend feature;
A structural defect matching feature set is determined based on the first defect state matching feature, the second defect state matching feature, the third defect state matching feature, and the fourth defect state matching feature.
3. The method of claim 1, wherein determining associated concrete BIM structure data corresponding to the target concrete BIM structure data from initial concrete BIM structure data of a cloud BIM structure information set according to the target BIM structure attribute vector, comprises:
Identifying a building scene tag to which the target concrete BIM structure data belongs;
Determining initial concrete BIM structure data under the building scene label in the cloud BIM structure information set;
And determining associated concrete BIM structure data corresponding to the target concrete BIM structure data from the initial concrete BIM structure data under the building scene label according to the target BIM structure attribute vector.
4. The method of claim 2, wherein the determining, based on the structural defect matching feature set and the associated concrete BIM structural data, a target structural defect detection result corresponding to the associated concrete BIM structural data comprises:
Determining target set defect labeling data corresponding to the associated concrete BIM structure data from the structure defect matching feature set;
When the target set defect labeling data comprises mechanical conduction data aiming at the associated concrete BIM structure data, processing the target concrete BIM structure data according to the mechanical conduction data to obtain a target structure defect detection result.
5. The method of claim 1, wherein the obtaining target concrete BIM structure data currently incoming by a detection interface of the target task server in the concrete structure defect detection system includes:
Generating at least one template concrete BIM structure data according to the building scene label of the target task server, and transmitting the template concrete BIM structure data to the target task server in a detection interface in the concrete structure defect detection system;
And selecting and processing the transmitted template concrete BIM structure data in the detection interface according to the target task server to acquire the target concrete BIM structure data currently transmitted by the target task server in the detection interface.
6. The method of claim 2, wherein the acquiring the target task server further comprises, after the target concrete BIM structure data currently entered by the detection interface in the concrete structure defect detection system:
Carrying out concrete BIM structure data quality analysis on the target concrete BIM structure data;
when the quality analysis report accords with a set index, transmitting concrete BIM structure data under a template BIM element to the target task server, wherein the template BIM element is different from BIM elements contained in the target concrete BIM structure data;
And when the quality analysis report does not accord with the set index, transmitting basic concrete BIM structure data under a target BIM element to the target task server, wherein the target BIM element is associated with BIM elements contained in the target concrete BIM structure data.
7. The method of claim 1, wherein the performing BIM structural attribute mining on the target concrete BIM structural data to obtain a target BIM structural attribute vector corresponding to the target concrete BIM structural data includes:
And performing BIM structure attribute mining on the target concrete BIM structure data through a BIM structure attribute mining algorithm to obtain a target BIM structure attribute vector corresponding to the target concrete BIM structure data.
8. The method of claim 7, wherein the performing BIM structure attribute mining on the target concrete BIM structure data by using a BIM structure attribute mining algorithm, before obtaining a target BIM structure attribute vector corresponding to the target concrete BIM structure data, further comprises:
Acquiring an algorithm sample, wherein the algorithm sample comprises a plurality of concrete BIM structure data examples;
Generating, for each concrete BIM structure data instance, a set of positive example tuples of the concrete BIM structure data instance, the set of positive example tuples comprising the concrete BIM structure data instance and at least one positive training data having a feature in common with the concrete BIM structure data instance;
sorting the concrete BIM structure data examples with each positive training data respectively to construct at least one first training example binary set of the concrete BIM structure data examples;
constructing at least one second training example binary group of the concrete BIM structure data example according to the concrete BIM structure data example and data in the positive example binary group set of each concrete BIM structure data example;
Performing BIM structure attribute mining on the data in the first training example binary group and the second training example binary group through a preset BIM structure attribute mining algorithm to obtain BIM structure attribute vectors of the data in the first training example binary group and the second training example binary group;
Determining attribute vector differences of the first training example binary group based on BIM structure attribute vectors of each content in the first training example binary group;
Determining attribute vector differences of the second training example binary group based on BIM structure attribute vectors of contents in the second training example binary group;
optimizing algorithm variables of a preset BIM structure attribute mining algorithm according to the attribute vector difference of the first training example binary group and the attribute vector difference of the second training example binary group to obtain an optimized BIM structure attribute mining algorithm;
The concrete BIM structure data example comprises a BIM element unit and a concrete space structure diagram example; the generating the positive example binary set of concrete BIM structure data examples includes: respectively carrying out image data derivatization on the BIM element unit and the concrete space structure diagram example to obtain a derivatization element unit of the BIM element unit and a concrete space structure derivatization diagram of the concrete space structure diagram example; determining a background image in the concrete space structure diagram example based on the set element word vector and a preset BIM element unit; updating the background image in the concrete space structure diagram example to obtain an updated concrete space structure diagram example; acquiring quantization characteristics of the BIM element unit; determining positive training data of the BIM element unit according to the derivative element unit and the quantization characteristic of the BIM element unit; determining positive training data of the concrete space structure diagram example according to the concrete space structure derivative diagram of the concrete space structure diagram example and the updated concrete space structure diagram example;
The optimizing the algorithm variable of the preset BIM structure attribute mining algorithm according to the attribute vector difference of the first training example binary group and the attribute vector difference of the second training example binary group to obtain an optimized BIM structure attribute mining algorithm comprises the following steps: integrating the attribute vector differences of each second training example binary group to obtain integrated second attribute vector differences; for each first training example binary group, determining local training cost information of the first training example binary group according to the integrated second attribute vector difference and the attribute vector difference of the first training example binary group; integrating the local training cost information of each first training example binary group to obtain global training cost information; and optimizing algorithm variables of a preset BIM structure attribute mining algorithm according to the global training cost information to obtain an optimized BIM structure attribute mining algorithm.
9. A concrete structure defect detection system, comprising: a memory for storing program instructions and data; a processor coupled to a memory for executing instructions in the memory to implement the method of any of claims 1-8.
10. A computer storage medium containing instructions which, when executed on a processor, implement the method of any of claims 1-8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410350098.7A CN117951546B (en) | 2024-03-26 | 2024-03-26 | BIM-combined concrete structure defect detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410350098.7A CN117951546B (en) | 2024-03-26 | 2024-03-26 | BIM-combined concrete structure defect detection method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117951546A CN117951546A (en) | 2024-04-30 |
CN117951546B true CN117951546B (en) | 2024-05-31 |
Family
ID=90803340
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410350098.7A Active CN117951546B (en) | 2024-03-26 | 2024-03-26 | BIM-combined concrete structure defect detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117951546B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112800512A (en) * | 2021-01-06 | 2021-05-14 | 常州工学院 | Safety detection method for reinforced concrete building structure |
CN116578568A (en) * | 2023-07-12 | 2023-08-11 | 八爪鱼人工智能科技(常熟)有限公司 | Data optimized storage method and big data system based on artificial intelligence |
CN117636073A (en) * | 2024-01-24 | 2024-03-01 | 贵州科筑创品建筑技术有限公司 | Concrete defect detection method, device and storage medium |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP4237939A4 (en) * | 2020-10-29 | 2024-07-17 | Veracode Inc | Development pipeline integrated ongoing learning for assisted code remediation |
-
2024
- 2024-03-26 CN CN202410350098.7A patent/CN117951546B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112800512A (en) * | 2021-01-06 | 2021-05-14 | 常州工学院 | Safety detection method for reinforced concrete building structure |
CN116578568A (en) * | 2023-07-12 | 2023-08-11 | 八爪鱼人工智能科技(常熟)有限公司 | Data optimized storage method and big data system based on artificial intelligence |
CN117636073A (en) * | 2024-01-24 | 2024-03-01 | 贵州科筑创品建筑技术有限公司 | Concrete defect detection method, device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN117951546A (en) | 2024-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10417524B2 (en) | Deep active learning method for civil infrastructure defect detection | |
CN115578015B (en) | Sewage treatment whole process supervision method, system and storage medium based on Internet of things | |
Xu et al. | Recognition of rust grade and rust ratio of steel structures based on ensembled convolutional neural network | |
CN110992307A (en) | Insulator positioning and identifying method and device based on YOLO | |
CN113590396A (en) | Method and system for diagnosing defect of primary device, electronic device and storage medium | |
CN115330268A (en) | Comprehensive emergency command method and system for dealing with mine disaster | |
Ali et al. | Wall crack detection using transfer learning-based CNN models | |
CN116861924A (en) | Project risk early warning method and system based on artificial intelligence | |
CN117789038B (en) | Training method of data processing and recognition model based on machine learning | |
Park et al. | A framework for improving object recognition of structural components in construction site photos using deep learning approaches | |
CN117951546B (en) | BIM-combined concrete structure defect detection method and system | |
CN113177644A (en) | Automatic modeling system based on word embedding and depth time sequence model | |
CN117422990A (en) | Bridge structure classification and evaluation method based on machine learning | |
CN112001484A (en) | Safety defect report prediction method based on multitask deep learning | |
CN116189010A (en) | Mine ecological identification early warning method and system based on satellite map | |
CN114897193A (en) | Airplane structure maintenance decision method and decision system based on man-in-the-loop | |
Debroy et al. | An apposite transfer-learned DCNN model for prediction of structural surface cracks under optimal threshold for class-imbalanced data | |
Fuchsberger et al. | A correlation network model for structural health monitoring and analyzing safety issues in civil infrastructures | |
Alshawabkeh et al. | Automated Pavement Crack Detection Using Deep Feature Selection and Whale Optimization Algorithm. | |
Pathak | Bridge health monitoring using CNN | |
CN117852155B (en) | Reinforcing steel bar construction state simulation method and system applied to artificial intelligence | |
Pan et al. | Sequential design command prediction using BIM event logs | |
Yani et al. | Plastic-type prediction based on digital image using multinomial Naïve Bayes method | |
CN116720786B (en) | KG and PLM fusion assembly quality stability prediction method, system and medium | |
CN117076454B (en) | Engineering quality acceptance form data structured storage method and system |
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 |