Building quality hidden danger analysis method and system based on cloud computing software

Abstract

The invention discloses a building quality hidden danger analysis method and system based on cloud computing software, relates to the field of building quality hidden danger analysis based on cloud computing software, is used for enhancing the early diagnosis and early warning capability of systematic and global major hidden dangers of a building structure, and comprises the following steps: obtaining the position of a defect; determining a first building component corresponding to the position of the defect in the building information model according to the position of the defect; verifying the first building component based on a pre-defined inter-component structural relationship rule set in the building information model; the inter-component structural relationship rule set is used for indicating structural relationships among different building components; associating the first building component with the defect under the condition that the verification of the first building component is successful; and under the condition that the verification of the first building component fails, determining a second building component corresponding to the defect according to the structural relationship rule set between the components, and associating the second building component with the defect.

Description

Technical Field

[0001] This invention relates to the field of building quality hazard analysis based on cloud computing software, and in particular to a method and system for building quality hazard analysis based on cloud computing software. Background Technology

[0002] Building Information Modeling (BIM) is a digital 3D model containing detailed information on all building components, recording the specific location, material, design load, and structural importance level of each beam, column, and wall. By matching the geographic coordinates of data collected on-site with the BIM model, the system can now accurately pinpoint the specific component where a crack in a photograph occurs. This allows analytical methods to incorporate component importance into risk assessments. For example, when a crack is detected on a main beam, the system immediately marks it as a high-risk event; while defects on decorative panels are marked as low-priority issues. This significantly improves the accuracy and effectiveness of alerts.

[0003] The core logic of this analytical method remains based on the assessment of individual, isolated defects. Even when combined with a BIM model, it only enhances the risk assessment of that "point." In actual engineering projects, some major structural hazards often do not manifest as a single, severe defect in their early stages. Instead, they are composed of multiple seemingly unrelated "minor issues" distributed in different locations, each with a low risk level. For example, a project's basement floor slab was found to have minor water seepage (classified as low risk by the system). Simultaneously, several hairline cracks appeared at the connection between the exterior wall infill and the frame beam on the third floor (also classified as low risk). Several windows on the fifth floor experienced difficulty opening and closing after installation; inspection revealed minor deformation of the window frames (classified as a general installation quality issue). These three issues were treated as three independent, low-risk events in the system. However, an experienced chief engineer, during a site inspection, connected these phenomena and astutely realized that this might not be an isolated quality problem, but rather caused by overall uneven settlement. This chain of problems, which is spatially dispersed and manifests in various forms and is triggered by a single root cause (foundation problems), is completely beyond the reach of existing analytical methods, thus missing the opportunity to provide early warning of this gradual and systemic major risk. Summary of the Invention

[0004] This invention provides a building quality hazard analysis method based on cloud computing software, which enhances the early diagnosis and warning capabilities for systemic and global major hazards in building structures.

[0005] In a first aspect, in order to solve the above-mentioned technical problems, the present invention provides a method for analyzing potential building quality hazards based on cloud computing software, including: obtaining the location of defects; Based on the location of the defect, determine the first building component corresponding to the location of the defect in the building information model; The first building component is validated based on a predefined set of structural relationship rules between components in the building information model; the set of structural relationship rules is used to indicate the structural relationships between different building components. If the first building component is successfully verified, the first building component will be associated with the defect. If the verification of the first building component fails, the second building component corresponding to the defect is determined according to the rule set of structural relationships between components, and the second building component is associated with the defect.

[0006] Optionally, based on the location of the defect, the first building component corresponding to the location of the defect is determined in the building information model, including: Obtain the first mapping relationship from the building information model; the first mapping relationship includes the mapping relationship between different building components and different locations; Determine the first building component corresponding to the location of the defect from the first mapping relationship.

[0007] Optionally, the method also includes: The morphological characteristics of the defects are matched with the typical failure modes of the components to obtain an inconsistency score; If the inconsistency score is less than the score threshold, the first building component is determined to have been successfully verified. If the inconsistency score is greater than or equal to the score threshold, the verification of the first building component is determined to be unsuccessful.

[0008] Optionally, typical failure modes of components include different defect types, morphological characteristics, and textual description keywords; morphological characteristics include one or more of the following: crack direction, width range, and water seepage location.

[0009] Optionally, determine inconsistency scores, including: Obtain the morphological characteristics and descriptive information of the defects; Determine the first similarity and the second similarity; the first similarity is the similarity between the morphological features of the defect and the failure mode of the corresponding first building component in the typical failure mode of the component; the second similarity is the similarity between the descriptive information of the defect and the failure description of the corresponding first building component in the typical failure mode of the component. Inconsistency scores are determined based on the first and second similarity scores.

[0010] Optionally, an inconsistency score is determined based on the first similarity and the second similarity, including: The inconsistency score is determined based on the inconsistency scoring formula, the first similarity score, and the second similarity score. The inconsistency scoring formula is as follows: F = w1 (1 - S1) + w2 (1 - S2) Where F represents the inconsistency score, w1 represents the first preset weight, w2 represents the second preset weight, S1 represents the first similarity, and S2 represents the second similarity.

[0011] Optionally, the second building component corresponding to the defect is determined based on the set of structural relationship rules between components, including: Based on the set of structural relationship rules between components, identify multiple components related to the first building component; Determine the inconsistency score for each relevant component; The component with the lowest inconsistency score is identified as the second building component corresponding to the defect.

[0012] Optionally, the multiple components associated with the first building component include: components that have a physical connection with the first building component and components that have a mechanical transmission relationship with the first building component.

[0013] Optionally, determining the first similarity and the second similarity includes: Based on the cosine similarity algorithm, the similarity between the morphological features of the defect and the failure mode of the corresponding first building component in the typical failure modes of the component is determined, and the first similarity is obtained. Based on the cosine similarity algorithm, the similarity between the description information of the defect and the failure description of the corresponding first building component in the typical failure mode of the component is determined; thus, the second similarity is obtained.

[0014] Secondly, the present invention provides a building quality hazard analysis system based on cloud computing software, the system comprising: The information acquisition module is used to locate the defect. The preliminary association module is used to determine the first building component corresponding to the location of the defect in the building information model based on the location of the defect; The rationality judgment module is used to verify the first building component based on the predefined set of structural relationship rules between components in the building information model; the set of structural relationship rules between components is used to indicate the structural relationship between different building components; The association module is used to associate the first building component with the defect if the first building component is successfully verified. The association module is also used to determine the second building component corresponding to the defect based on the structural relationship rule set between components when the first building component fails to be verified, and to associate the second building component with the defect.

[0015] Compared with the prior art, the present invention has the following beneficial effects: This application discloses a method for analyzing potential building quality hazards based on cloud computing software. The method obtains the location of the defect and identifies the corresponding first building component in a building information model (BIM). Then, it verifies the component based on a predefined set of structural relationship rules between components. If verification is successful, the first building component is associated with the defect; if verification fails, a second building component corresponding to the defect is determined according to the set of structural relationship rules and associated with it.

[0016] Through the above technical solution, this application overcomes the limitations of existing technologies that only focus on assessing single, isolated defect points, effectively solving the problems of insufficient understanding of the defect's environment, susceptibility to "alarm fatigue," and missed opportunities for early warning of systemic major risks. By introducing a set of structural relationship rules between components, this application achieves an upgrade in analysis from "point" to "surface," enabling the identification of deep, systemic hidden dangers composed of multiple seemingly unrelated but actually interconnected "small problems." For example, when minor water seepage, hairline cracks, and minor window frame deformations are systematically linked to structural problems such as uneven settlement, this application can provide earlier and more accurate warnings, significantly improving the accuracy and effectiveness of building quality hazard analysis, avoiding invalid or low-priority alarms, and providing more reliable protection for engineering quality and structural safety. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of a building quality hazard analysis method based on cloud computing software provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of another method for analyzing potential building quality hazards based on cloud computing software provided in this embodiment of the invention; Figure 3 This is a schematic diagram of a building quality hazard analysis system based on cloud computing software provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0019] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0020] Traditional building quality management systems are inadequate in identifying deep-seated, systemic risks when dealing with complex, non-isolated building quality hazards. These systems often focus on checking and issuing warnings for single, clearly defined quantitative indicators. They fall short when dealing with difficult-to-quantify, subjectively judged phenomena, especially when these phenomena are not isolated but rather comprised of multiple seemingly unrelated but actually interconnected "small problems." Furthermore, even with the introduction of image and text processing modules and the integration of Building Information Modeling (BIM), the core logic of existing methods remains based on the assessment of individual, isolated defects, failing to effectively identify systemic major risks composed of multiple seemingly unrelated "small problems" distributed in different locations and with low inherent risk levels.

[0021] The following specific embodiments will provide a detailed introduction and explanation of a building quality hazard analysis method based on cloud computing software provided in this application.

[0022] Reference Figure 1 This invention provides a method for analyzing potential building quality hazards based on cloud computing software, comprising the following steps: S1, locate the defect.

[0023] The location of a defect refers to the geographical or spatial coordinates of any abnormal phenomena discovered during the construction or operation of a building that may affect the quality or safety of the building.

[0024] One possible approach is to obtain the location of the defect using on-site survey equipment (such as GPS or laser scanners) or by manually inputting a relative location description.

[0025] For example, on-site inspectors used mobile devices to photograph a wall crack and recorded its GPS coordinates. This location information was then uploaded to a cloud computing platform.

[0026] S2. Based on the location of the defect, determine the first building component corresponding to the location of the defect in the building information model.

[0027] Building Information Modeling (BIM) is a digital model that integrates all relevant information about a building project, including geometric information, component attributes, and structural relationships. The "structural relationship rule set between components" in a BIM model is a predefined set of rules used to describe the physical connections, mechanical transmission, and functional dependencies between different building components. For example, there is a load-bearing relationship between beams and columns, and a support relationship between walls and floor slabs. These rules can be stored in the BIM model in the form of logical expressions, diagrams, or databases to guide the verification and reasoning of relationships between components.

[0028] As one possible implementation, a first mapping relationship can be obtained from the building information model, and the first building component corresponding to the location of the defect can be determined from the first mapping relationship.

[0029] It should be noted that the first mapping relationship includes the mapping relationship between different building components and different locations.

[0030] The rule set for structural relationships between components provides a detailed explanation of the physical connections, mechanical transmission, and functional dependencies between different building components: Physical Connections: These rules describe the direct, visible connections between building components. For example, in a BIM model, rules can be defined such as "beams and columns are connected by rigid or hinged joints," "floor slabs and beams are connected by concrete pouring to form a whole," and "there is direct support between walls and foundations." These rules clarify which components are spatially adjacent and directly connected, collectively forming local structural units of the building. For instance, if a defect is initially associated with a beam, the system will identify the columns and floor slabs directly connected to that beam as potential related components based on the physical connection rules.

[0031] Mechanical transmission relationships: These rules describe how loads and deformations are transferred between building components. This relationship doesn't necessarily require a direct physical connection, but is based on structural mechanics principles. For example, a set of rules might include statements such as "a uniformly distributed load on a floor slab will be transferred to the beams supporting it," "a load on a beam will be transferred to the columns at its ends," and "a load on a column will be transferred down to the foundation layer by layer." These rules are crucial for identifying the underlying causes of defects. For instance, a crack in a non-load-bearing wall on the top floor might not be a problem with the wall itself, but rather caused by uneven settlement of the load-bearing beams or foundation below, leading to deformation of the superstructure through mechanical transmission relationships.

[0032] These rules can be stored and represented in various forms within the BIM model, such as logical expressions, graphical structures (like knowledge graphs), or relational tables in a database. They provide the system with knowledge about "how" the building structure "works," enabling the system to reason and make judgments.

[0033] S3. Verify the first building component based on the predefined set of structural relationship rules between components in the building information model.

[0034] Among them, the structural relationship rule set between components is used to indicate the structural relationship between different building components.

[0035] As one possible approach, it is possible to check whether the defect conforms to the typical failure mode of the first building component.

[0036] If the defect matches the typical failure mode of the first building component, the first building component is deemed to have been successfully verified.

[0037] If the defect does not conform to the typical failure mode of the first building component, the verification of the first building component is determined to be unsuccessful.

[0038] For example, if the defect is a crack and the first building component is a wall component, and the direction and width range of the crack do not conform to the typical failure mode of the wall component (e.g., it is outside the width range set in the typical failure mode, or the direction is different from the setting in the typical failure mode), then the verification of the first building component is determined to be unsuccessful.

[0039] For example, if the defect is a crack and the first building component is a wall component, if the direction and width of the crack match the typical failure mode of the wall component (e.g., it is within the setting range of the typical failure mode and has the same direction as the setting in the typical failure mode), then the first building component is determined to have been successfully verified.

[0040] S4. If the first building component is successfully verified, associate the first building component with the defect.

[0041] For example, if the system determines that the wall component is indeed likely to develop such cracks, and the crack characteristics match the typical failure mode of the component, then the crack defect is directly associated with the wall component and recorded in the database.

[0042] S5. If the verification of the first building component fails, determine the second building component corresponding to the defect according to the structural relationship rule set between components, and associate the second building component with the defect.

[0043] Based on the set of structural relationship rules between components, identify multiple components related to the first building component; and check whether the defect conforms to the typical failure modes of multiple components related to the first building component.

[0044] If the typical failure mode of the target component among the multiple components related to the first building component matches the defect, then the target component is identified as the second building component corresponding to the defect.

[0045] It should be noted that the verification method for the typical failure mode of the target component among the multiple components related to the defect and the first building component to match the defect is the same as the verification method described above, and will not be repeated here.

[0046] Compared with existing technologies, the advantage of this application lies in its ability to overcome the limitations of single components. Through in-depth analysis of the structural relationships between components, it achieves a more accurate and comprehensive identification of potential building quality hazards. Traditional methods often fall short when faced with systemic hazards composed of multiple seemingly unrelated but actually interconnected "small problems." This application, by introducing a set of rules governing structural relationships between components, enables the system to perform deeper reasoning and judgment, linking surface defects with deep structural issues, thereby effectively avoiding "alarm fatigue" and the underreporting of major hazards. This "verification-re-correlation" mechanism makes this method significantly progressive in identifying progressive, systemic major risks, enabling earlier detection and warning of potential structural problems, and providing a more intelligent and reliable solution for the quality management of building projects.

[0047] In one possible design, such as Figure 2 As shown, in order to verify the first building component, this application may further include the following steps: S101. Match the morphological characteristics of the defects with the typical failure modes of the components to obtain an inconsistency score.

[0048] Among them, the typical failure modes of components include the defect types, morphological characteristics and textual description keywords of different components; the morphological characteristics include one or more of the following: crack direction, width range and water seepage location.

[0049] One possible approach is to obtain the morphological features and descriptive information of the defect, determine the first similarity and the second similarity, and determine the inconsistency score based on the first similarity and the second similarity.

[0050] Specifically, when obtaining the morphological characteristics and descriptive information of defects, they can be extracted from on-site survey data, image recognition results, or manually entered text. Morphological characteristics can include intuitive geometric or physical manifestations such as crack direction, width range, and seepage location, while descriptive information can be a textual description of the defect, such as "wall cracking" or "floor slab seepage."

[0051] The first similarity is the similarity between the morphological features of the defect and the failure mode of the corresponding first building component in the typical failure mode of the component; the second similarity is the similarity between the descriptive information of the defect and the failure description of the corresponding first building component in the typical failure mode of the component. For example, an inconsistency score can be determined based on the inconsistency scoring formula, the first similarity score, and the second similarity score. The inconsistency scoring formula is: F = w1 (1 - S1) + w2 (1 - S2) Where F represents the inconsistency score, w1 represents the first preset weight, w2 represents the second preset weight, S1 represents the first similarity, and S2 represents the second similarity. The first preset weight w1 and the second preset weight w2 can be set according to expert experience.

[0052] S102. If the inconsistency score is less than the score threshold, the first building component is determined to have been successfully verified.

[0053] The scoring threshold can be set as needed and is not restricted.

[0054] S103. If the inconsistency score is greater than or equal to the score threshold, the verification of the first building component is determined to be unsuccessful.

[0055] This application's solution introduces a structured inconsistency scoring formula, comprehensively considering the degree of matching between the morphological characteristics of the defect and the typical failure mode of the component, as well as the degree of matching between the descriptive information of the defect and the typical failure mode of the component, through a weighted summation. Specifically, similarity is converted into dissimilarity by calculating (1 - S1) and (1 - S2), allowing the inconsistency score F to intuitively reflect the difference between the defect and the typical failure mode of the component. The introduction of the first preset weight w1 and the second preset weight w2 allows the system to flexibly adjust the relative importance of morphological characteristics and descriptive information in evaluating the component verification results based on actual needs or expert experience. Thus, this formula provides a quantitative and adjustable index for judging whether the first building component matches the actual situation of the defect, thereby effectively supporting the subsequent judgment of the success or failure of component verification.

[0056] In some preferred embodiments, a specific example is given below. Suppose a crack is found in a wall of a building, and its location is initially identified as corresponding to a "load-bearing wall" in the Building Information Model (BIM). To verify whether the "load-bearing wall" is indeed the component from which the crack occurred, the system obtains the crack's morphological characteristics, such as its oblique direction, width of 2 mm, and length of 1.5 m. Simultaneously, the system queries a database of typical failure modes for "load-bearing walls," which may record common crack types for "load-bearing walls." For example, the typical characteristics of a "stress crack" are vertical or oblique, with a width typically between 0.5 and 1 mm, while the typical characteristics of a "settlement crack" are oblique, with potentially larger and uneven widths. The system matches the crack's morphological characteristics with the typical failure modes of the "load-bearing wall" and calculates an inconsistency score. For example, if the crack's morphological characteristics highly match the "stress crack" mode of the "load-bearing wall," the inconsistency score might be 0.2. If the preset scoring threshold is 0.5, since 0.2 is less than 0.5, the system will determine that the "load-bearing wall" verification is successful and associate the crack with the "load-bearing wall". Conversely, if the morphological characteristics of the crack do not match any typical failure mode of the "load-bearing wall", for example, if the inconsistency score is calculated as 0.7, which is greater than the scoring threshold of 0.5, the system will determine that the "load-bearing wall" verification fails and initiate subsequent processes to find a second building component that better matches the defect.

[0057] Through the above technical solution, this application provides a quantitative and flexible method for calculating inconsistency scores. This method comprehensively considers the morphological characteristics and descriptive information of defects and allows for adjustments to preset weights w1 and w2 to adapt to different evaluation focuses, thereby making the calculation of inconsistency scores more accurate and intelligent. This weighted summation calculation method not only improves the accuracy of component verification but also enhances the system's adaptability and robustness when handling complex and variable building quality hazard data, providing a more reliable basis for subsequent defect correlation.

[0058] Traditional methods for analyzing building quality defects, after initially locating and associating defects with a first building component, require identifying a second building component if the verification of that first component fails. However, existing technologies lack specific and systematic guidance on how to accurately and effectively determine this second building component, potentially leading to arbitrary or inaccurate selection and affecting the reliability of defect association. Failure to address this issue may result in incorrect association between defect information and the actual damaged component, impacting subsequent maintenance decisions and quality management efficiency. Therefore, this application proposes a more accurate and robust method for determining the second building component, aiming to improve the accuracy of defect association through a comprehensive consideration of structural relationships and inconsistency scoring.

[0059] In this regard, this application further proposes that the above-mentioned method also includes: S201. Based on the set of structural relationship rules between components, determine multiple components related to the first building component.

[0060] Among them, the multiple components related to the first building component include: components that have a physical connection with the first building component and components that have a mechanical transmission relationship with the first building component.

[0061] As one possible implementation, when the first building component fails to be verified, the system no longer directly associates the defect with the failed component. Instead, it uses the predefined set of structural relationship rules between components to pre-store the physical connection relationships and mechanical transmission relationships between different components to identify multiple components that are physically or mechanically related to the first building component.

[0062] S202. Determine the inconsistency score for each relevant component.

[0063] The inconsistency score is an indicator that measures the degree of matching between components and defects; the lower the score, the higher the degree of matching.

[0064] The inconsistency score for each relevant component is determined in the same way as the inconsistency score for the first building component, and will not be described in detail here.

[0065] S203. The component with the lowest inconsistency score is identified as the second building component corresponding to the defect.

[0066] This application's solution effectively solves the problem of accurately determining the second building component when the verification of the first building component fails by introducing a comprehensive judgment mechanism of structural relationship rules between components and inconsistency scoring. Specifically, when the initially associated first building component is determined to be mismatched, the system does not blindly re-search, but first uses the structural relationship rules between components to limit the search scope to the set of components that are structurally related to the first building component. This limitation based on structural relationships ensures that the selected candidate components are physically and mechanically reasonable to the location of the defect, avoiding interference from irrelevant components. On this basis, by calculating the inconsistency score between each related component and the defect, a quantitative assessment of the matching degree between the candidate component and the defect is achieved. The introduction of inconsistency scoring enables the system to make a refined judgment on the correlation between components and defects from multiple dimensions (such as morphological features, descriptive information, etc.). Finally, the component with the lowest inconsistency score is selected as the second building component, ensuring that the determined second building component is the component with the highest correlation to the defect and most consistent with the actual situation. It is precisely because of this mechanism that combines structural relationship constraints with quantitative matching degree assessment that the association between defects and building components becomes more accurate and reliable.

[0067] Through the above technical solution, this application provides a systematic and intelligent method to determine the second building component corresponding to the defect when the verification of the first building component fails. Compared with judgment based solely on location information or simple rules, this solution significantly improves the accuracy and reliability of associating defects with actual damaged building components by considering the structural relationships between components and combining inconsistency scoring to quantitatively evaluate the relevant components. This helps avoid erroneous associations caused by initial judgment errors, thereby improving the overall accuracy of building quality hazard analysis and providing a more solid data foundation for subsequent hazard investigation, maintenance decisions, and quality management, effectively reducing potential engineering risks and maintenance costs.

[0068] In some preferred embodiments, a specific example is given below. Suppose that in a building information model, a crack defect located on the surface of a wall is initially identified. Based on the location of the defect, the system first associates it with a first building component, namely the wall. However, when verifying the wall, it is found that the morphological characteristics of the crack (e.g., crack direction, width) differ significantly from the typical failure modes of the wall (e.g., shear cracks, shrinkage cracks), causing the wall verification to fail.

[0069] At this point, the proposed solution is initiated. First, based on the structural relationship rule set between components, the system identifies multiple components related to the wall. For example, beams, columns, and floor slabs adjacent to the wall are identified as related components. Next, the system calculates the inconsistency scores of the beams, columns, and floor slabs relative to the crack defect. Assuming that the calculated inconsistency score is 0.2 for the beam, 0.6 for the column, and 0.4 for the floor slab, and since the beam has the lowest inconsistency score, the system identifies the beam as the second building component corresponding to the crack defect. In this way, even if the wall is initially judged to be mismatched, the system can accurately associate the crack defect with the potentially damaged beam by considering structural relationships and quantifying the matching degree, thereby avoiding misjudgment and improving the accuracy of defect association.

[0070] In one possible design, to determine the first similarity and the second similarity, this application further includes the following steps: S301. Based on the cosine similarity algorithm, determine the similarity between the morphological characteristics of the defect and the failure mode of the corresponding first building component in the typical failure mode of the component, and obtain the first similarity.

[0071] S302. Based on the cosine similarity algorithm, determine the similarity between the description information of the defect and the failure description of the corresponding first building component in the typical failure mode of the component; and obtain the second similarity.

[0072] Specifically, cosine similarity is a metric that measures the cosine of the angle between two non-zero vectors, with a value between -1 and 1. The closer the directions of two vectors are, the larger the cosine similarity value, indicating that they are more similar. The cosine similarity algorithm obtains the cosine value by calculating the ratio of the dot product of two vectors to the product of their respective moduli.

[0073] In practical applications, the morphological characteristics and descriptive information of defects, as well as the corresponding failure modes and failure descriptions in typical component failure modes, can all be converted into numerical feature vectors.

[0074] For example, morphological features can include numerical values ​​such as the length, width, and depth of a crack, which can form a morphological feature vector. Descriptive information can be converted into semantic vectors using natural language processing (NLP) techniques, such as word embeddings or sentence embeddings. By representing these features as vectors, the cosine similarity algorithm can be used to efficiently calculate the degree of similarity between them.

[0075] The specific calculation process of the cosine similarity algorithm can be found in existing technologies and will not be elaborated here.

[0076] The proposed solution uses cosine similarity of feature vectors to determine the first and second similarities, enabling a quantitative and accurate assessment of the matching degree between defect features and typical component failure modes. The cosine similarity algorithm effectively handles high-dimensional data and is insensitive to vector length, focusing only on the vector direction, i.e., the relative distribution of features. Therefore, even if the absolute values ​​of the morphological features or descriptive information of defects differ, they can be accurately identified as long as their intrinsic patterns or semantic content are similar. This helps improve the accuracy and robustness of similarity calculations, providing reliable input for subsequent inconsistency scoring.

[0077] By employing the aforementioned technical solution and using a cosine similarity algorithm based on feature vectors to determine the first and second similarities, precise quantification of the similarity between the morphological features and descriptive information of defects and typical failure modes of components can be achieved. This method not only improves the accuracy and objectivity of similarity calculation, avoiding errors caused by subjective judgment, but also enhances the system's ability to process complex and diverse defect data. Therefore, the verification results of the first building component can be evaluated more accurately, thereby improving the reliability and efficiency of building quality hazard analysis.

[0078] like Figure 3 As shown in the figure, this invention also provides a building quality hazard analysis system based on cloud computing software. The system includes: The information acquisition module is used to locate the defect. The preliminary association module is used to determine the first building component corresponding to the location of the defect in the building information model based on the location of the defect; The rationality judgment module is used to verify the first building component based on the predefined set of structural relationship rules between components in the building information model; the set of structural relationship rules between components is used to indicate the structural relationship between different building components; The association module is used to associate the first building component with the defect if the first building component is successfully verified. The association module is also used to determine the second building component corresponding to the defect based on the structural relationship rule set between components when the first building component fails to be verified, and to associate the second building component with the defect.

[0079] This application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be executed by a computer program instructing related hardware. This program can be stored in the computer-readable storage medium, and when executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be an internal storage unit of the task execution device (including a data sending end and / or a data receiving end) of any of the foregoing embodiments, such as the hard disk or memory of the task execution device. The computer-readable storage medium can also be an external storage device of the terminal device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the terminal device. Further, the computer-readable storage medium can include both the internal storage unit of the task execution device and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by the task execution device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0080] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0081] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, essentially, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0082] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A cloud computing software-based building quality hidden danger analysis method, characterized in that, The method comprises: obtaining a position of a defect; determining a first building component corresponding to the position of the defect in a building information model according to the position of the defect; verifying the first building component based on a predefined set of inter-component structural relationship rules in the building information model; the set of inter-component structural relationship rules are used to indicate structural relationships between different building components; in a case where the first building component is verified successfully, associating the first building component with the defect; in a case where the first building component is verified unsuccessfully, determining a second building component corresponding to the defect according to the set of inter-component structural relationship rules, and associating the second building component with the defect.

2. The building quality hidden danger analysis method based on cloud computing software according to claim 1, characterized in that, The method comprises: obtaining a first mapping relationship from the building information model; the first mapping relationship comprises mapping relationships between different building components and different positions; determining the first building component corresponding to the position of the defect from the first mapping relationship. 3.The building quality hidden danger analysis method based on cloud computing software according to claim 2, characterized in that, The method further comprises: matching a morphological feature of the defect with a typical failure mode of a component to obtain an inconsistency score; in a case where the inconsistency score is less than a score threshold, determining that the first building component is verified successfully; in a case where the inconsistency score is greater than or equal to the score threshold, determining that the first building component is verified unsuccessfully.

4. The building quality hidden danger analysis method based on cloud computing software according to claim 3, characterized in that, The typical failure mode of the component comprises a defect type, a morphological feature, and a text description keyword of different components; the morphological feature comprises one or more of a crack direction, a width range, and a water seepage position.

5. The building quality hidden trouble analysis method based on cloud computing software according to claim 3, characterized in that, The method comprises: obtaining the morphological feature and description information of the defect; determining a first similarity and a second similarity; the first similarity is a similarity between the morphological feature of the defect and a failure mode corresponding to the first building component in the typical failure mode of the component; the second similarity is a similarity between the description information of the defect and a failure description corresponding to the first building component in the typical failure mode of the component; determining the inconsistency score according to the first similarity and the second similarity.

6. The building quality hidden trouble analysis method based on cloud computing software according to claim 5, characterized in that, The method comprises: determining the inconsistency score according to an inconsistency score formula, the first similarity, and the second similarity; the inconsistency score formula is: F = w1 (1 - S1) + w2 (1 - S2) wherein F represents the inconsistency score, w1 represents a first preset weight, w2 represents a second preset weight, S1 represents the first similarity, and S2 represents the second similarity.

7. The building quality hidden trouble analysis method based on cloud computing software according to claim 3, characterized in that, The method comprises: determining a plurality of components related to the first building component according to the set of inter-component structural relationship rules; determining an inconsistency score of each related component; determining the second building component corresponding to the defect as a related component with the lowest inconsistency score. 8.The building quality hidden danger analysis method based on cloud computing software according to claim 7, wherein, The plurality of components related to the first building component include components having a physical connection relationship with the first building component and components having a mechanical conduction relationship with the first building component. 9.The building quality hidden danger analysis method based on cloud computing software according to claim 5, wherein, The determining the first similarity and the second similarity comprises: According to a cosine similarity algorithm, determining a similarity between a morphological feature of the defect and a failure mode corresponding to the first building component in a typical failure mode of the component, to obtain the first similarity; According to the cosine similarity algorithm, determining a similarity between description information of the defect and a failure description corresponding to the first building component in a typical failure mode of the component, to obtain the second similarity.

10. A cloud computing software-based building quality hidden danger analysis system, characterized in that, The system comprises: An information acquisition module configured to acquire a position of a defect; A preliminary association module configured to determine a first building component corresponding to the position of the defect in a building information model according to the position of the defect; A rationality judgment module configured to verify the first building component based on a pre-defined component structure relationship rule set in the building information model; the component structure relationship rule set is used to indicate a structural relationship between different building components; An association module configured to associate the first building component with the defect when the first building component is verified successfully; The association module is further configured to determine a second building component corresponding to the defect according to the component structure relationship rule set when the first building component fails to be verified, and to associate the second building component with the defect.

Images