A disease attribute determination and scale preliminary evaluation method and system

Abstract

The application discloses a disease attribute judgment and scale preliminary evaluation method and system, and the method comprises the following steps: acquiring bridge detection data, binding to specific structural members after identifying the disease type, determining the spatial distribution boundary of the disease on the member, and synchronously recording the structure part and key stress attribute to which the member belongs; based on the member binding information, the full-dimensional attribute parameters of the disease are classified and extracted, and the standardized disease attribute parameter set is formed; an industry standard structured knowledge base is established, the standardized parameter set is automatically matched with the scale judgment condition of the corresponding disease in the library, the candidate scale grades meeting the conditions are screened, and the multi-scale candidate set is constructed; a preset engineering experience threshold library is called, the structure safety influence corresponding to the candidate scale is evaluated, the lowest value of the candidate scale is taken if the safety is not affected, the importance level of the member is adjusted if there is a risk, and finally the disease scale preliminary evaluation result and judgment basis are output. The long-term safe and stable operation of the bridge structure is ensured.

Description

Technical Field

[0001] This invention belongs to the field of bridge engineering defect detection and assessment technology, and more specifically, relates to a defect attribute determination and scale preliminary assessment method and system. Background Technology

[0002] In the field of bridge engineering, the determination of bridge defects and their scaling assessment are crucial steps in ensuring bridge structural safety and making maintenance decisions. For a long time, bridge defect detection has relied heavily on manual visual inspection or single detection equipment. The resulting data is diverse and lacks effective integration, leading to vague defect localization and difficulty in accurately associating defects with specific bridge components and their structural properties. For example, inspection images can only present the superficial characteristics of defects but cannot clearly define their specific location within the bridge's three-dimensional structure, nor can they intuitively reflect the stress importance of the associated components. This lack of accurate spatial and structural context makes subsequent attribute analysis and scaling determination prone to misjudgment.

[0003] Meanwhile, the quantification of defect attributes lacks a unified standard. Data formats such as images and point clouds obtained through different detection methods vary, and the methods for extracting attribute parameters also differ, making it difficult to standardize the analysis and comparison of attribute data. Traditional scaling judgments rely heavily on human experience and subjective judgment based on industry standards. This is not only inefficient but also easily influenced by individual differences in experience, making it difficult to guarantee the objectivity and consistency of the judgment results. Furthermore, the scaling judgment process often overlooks the correlation between structural safety risks and the importance of components. For defects in critical load-bearing components, if the impact on structural safety cannot be accurately assessed, the risk is easily underestimated, creating hidden dangers for bridge structural safety.

[0004] These problems hinder the precise and intelligent management of bridge defects, leading to a lack of scientific basis for maintenance decisions. This can result in wasted maintenance resources and potential structural safety accidents due to misjudgments of defect risks. Therefore, there is an urgent need for a method that can accurately locate defects, standardize and quantify attributes, and objectively and intelligently determine their severity. This method would address the current problems in bridge defect management, such as vague location, inconsistent attribute quantification, subjective and inefficient scaling, and a lack of safety risk considerations. Ultimately, it would improve the scientific rigor and efficiency of bridge defect management, ensuring the long-term safe and stable operation of bridge structures. Summary of the Invention

[0005] This invention aims to provide a method for determining the attributes of bridge defects and for preliminary scaling assessment. This method enables precise spatial positioning of defects on bridge components, standardizes and quantifies the attributes of defects, and automatically matches scaling based on industry standards and engineering experience to assess the impact on structural safety. It solves the problems of vague positioning, inconsistent attribute quantification, and subjective and inefficient scaling assessment in traditional methods, and provides scientific support for bridge defect management and maintenance decisions.

[0006] To address the aforementioned deficiencies or improvement needs of existing technologies, as a first aspect of this invention, the present invention provides a method for determining disease attributes and preliminary scaling, comprising:

[0007] S1. Obtain bridge inspection data, identify the type of defects, bind the defects to specific structural components of the bridge, clarify the spatial distribution boundary of the defects on the component, and record the structural part of the bridge to which the component belongs and its key stress attributes.

[0008] S2. Based on the bound component information, classify and extract the full-dimensional attribute parameters of the disease to form a standardized set of disease attribute parameters;

[0009] S3. Establish a structured knowledge base with industry standards; automatically match the obtained disease attribute parameter set with the scaling judgment conditions of the corresponding disease type in the knowledge base, and filter out all qualified candidate scaling levels to form a multi-scale candidate set;

[0010] S4. Based on the identification of disease risks, scale selection is performed on the multi-scale candidate set; if the assessment confirms that the disease does not affect structural safety, the lowest value among the candidate scales is taken as the initial assessment scale; if there is a structural safety risk, the highest value among the candidate scales is taken as the initial assessment scale, and finally the initial assessment results of the disease scale and the judgment basis are output.

[0011] Furthermore, the types of defects in S1 include: cracks, spalling and exposed reinforcement, wear, deformation, water damage, corrosion, foundation erosion, pavement damage, and other bridge structure-related defects.

[0012] Furthermore, the method for determining the spatial distribution boundary of the defects on the component in S1 is as follows:

[0013] Based on bridge inspection data, three-dimensional spatial data and image data of the components where the defects are located are obtained. The three-dimensional spatial data includes the three-dimensional coordinate information of the components and defects, and the image data is used to locate the two-dimensional area range of the defects. The region feature set of the defects in the image is extracted by the image segmentation method, and the mapping relationship between the region feature set and the three-dimensional spatial data is established, converting the defect region features in the two-dimensional image into the corresponding three-dimensional defect point set.

[0014] The three-dimensional defect point set is optimized to eliminate redundant data, and then a targeted search space is determined based on the geometric features of the three-dimensional defect point set. Candidate component surfaces are retrieved in the search space. By calculating the distance from representative points in the three-dimensional defect point set to each candidate component surface, the component surface with the best matching degree is selected as the binding component of the defect, thus completing the accurate association between the defect and the component.

[0015] The surface type of the component is identified based on the surface features of the bound component. Then, a local coordinate system is established with the surface of the bound component as the reference. The three-dimensional disease point set is projected onto the two-dimensional plane of the local coordinate system to obtain the two-dimensional disease point set. According to the geometric morphological characteristics of the disease, the spatial distribution boundary of the disease is extracted from the two-dimensional disease point set using the corresponding boundary extraction algorithm. This boundary is used for subsequent calculation of disease attribute parameters and scaling evaluation.

[0016] Furthermore, the full-dimensional attribute parameters in S2 include spatial attributes, geometric attributes, material attributes, and environmental related attributes.

[0017] Furthermore, the process of constructing the standardized disease attribute parameter set in S2 is as follows:

[0018] For planar defects, spatial and geometric attributes are extracted based on the convex hull closed boundary curve of the defect. The spatial attributes include location information determined by the boundary center point; the geometric attributes include the defect area calculated using a surface integral algorithm and the defect depth determined by statistical values ​​of the distance from the defect point cloud to the component surface.

[0019] For linear defects, spatial and geometric attributes are extracted based on the defect skeleton line and width distribution. The spatial attributes include the orientation determined by the skeleton line direction, the position determined by the boundary center point, and the distribution pattern determined by the skeleton line length and width distribution. The geometric attributes include the skeleton line length calculated using the line integral algorithm, the overall width of the defect obtained based on the skeleton line normal traversal, and the defect depth determined by the maximum distance from the defect point cloud to the component surface.

[0020] Then, based on the distance between the detection point and the center point of the disease boundary, the valid detection data are screened, and the statistical mean of the valid data is used as the quantitative value of the disease material properties.

[0021] Then, environmental correlation attributes are obtained through external data interfaces, and environmental corrosion factors are constructed using a normalized weighted model. Operational load impact factors are constructed based on the Miner damage principle.

[0022] The spatial attributes, geometric attributes, material attributes, and environmental related attributes are organized according to a preset format to form a standardized set of disease attribute parameters.

[0023] Furthermore, the knowledge base in S3 contains scaling conditions corresponding to various diseases.

[0024] Furthermore, the construction process of the multi-scale candidate set in S3 is as follows:

[0025] Based on a standardized set of disease attribute parameters, quantitative values ​​of each attribute are extracted and matched with scaling conditions for the corresponding disease type in a structured knowledge base. The matching includes quantitative condition matching and qualitative condition matching.

[0026] Quantitative condition matching takes geometric parameters and material property parameters as the core, calculates the proportion of diseased area and crack length ratio according to industry standards, compares the material property parameters with standard thresholds, and maps the calculation and comparison results to the corresponding scaling level.

[0027] Qualitative condition matching is based on the spatial attributes and apparent characteristics of the disease, and the corresponding scale level is obtained by comparing it with the qualitative description judgment criteria in the knowledge base.

[0028] All scale levels obtained from quantitative and qualitative condition matching are integrated to form a multi-scale candidate set.

[0029] Furthermore, the confirmation process for the initial assessment results of the disease scaling in S4 is as follows:

[0030] For the multiple candidate scales formed, scale selection is carried out based on the identification of the nature of the disease; this process relies on the key stress attributes of the components and the identification results of the disease type recorded in the component binding process to complete the disease nature discrimination;

[0031] When the disease belongs to the superficial disease category and the component where the disease is located is a non-critical load-bearing component, the conservative scaling mapping mechanism is activated, and the minimum value in the multi-scale candidate set is selected as the initial evaluation scale; when the disease belongs to the structural disease category or the component where the disease is located is a critical load-bearing component, the safety priority scaling selection mechanism is activated, and the maximum value in the multi-scale candidate set is selected as the initial evaluation scale.

[0032] The determination of the nature of the disease needs to be based on the correlation analysis between the key stress attributes of the component and the spatial distribution characteristics of the disease. The key stress components are automatically identified through their structural parts and stress characteristics.

[0033] Finally, the initial assessment results of the disease scaling are output, and a complete set of judgment criteria including the basis for disease nature identification, component location correlation analysis, and scale optimization path are generated;

[0034] At the same time, the system fully records a standardized set of disease attribute parameters, which covers material properties and environmental related attributes, providing a detailed data foundation for subsequent disease development trend judgment and mechanism analysis.

[0035] As a second aspect of the present invention, a disease attribute determination and scaling preliminary assessment system is also provided, comprising:

[0036] The component binding and spatial boundary definition unit is used to acquire bridge inspection data, identify the type of defects, bind the defects to specific structural components of the bridge, clarify the spatial distribution boundary of the defects on the component, and record the bridge structural part and key stress attributes to which the component belongs.

[0037] The full-dimensional attribute parameter extraction and standardization unit is used to extract full-dimensional attribute parameters of diseases based on the bound component information, forming a standardized set of disease attribute parameters;

[0038] The standard matching and multi-scale candidate set construction unit is used to establish a structured knowledge base of industry standards; the obtained disease attribute parameter set is automatically matched with the scaling judgment conditions of the corresponding disease type in the knowledge base, and all qualified candidate scaling levels are selected to form a multi-scale candidate set;

[0039] The safety impact assessment and scale determination unit is used to select scales from a multi-scale candidate set based on the identification of disease risks. If the assessment confirms that the disease does not affect structural safety, the lowest value among the candidate scales is taken as the initial assessment scale. If there is a structural safety risk, the highest value among the candidate scales is taken as the initial assessment scale. Finally, the initial assessment results of the disease scale and the judgment basis are output.

[0040] As a third aspect of the present invention, a computer-readable storage medium is also provided, on which a computer program is stored, the computer program being executed by a processor of any of the methods described in the present invention for determining disease attributes and performing initial scaling.

[0041] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0042] 1. This invention provides a method for determining the attributes of bridge defects and for preliminary scaling assessment. By acquiring bridge inspection data, identifying defect types, and then binding the defects to specific structural components of the bridge, the spatial distribution boundaries of the defects on those components are clarified. Simultaneously, the structural location and key stress attributes of the components are recorded, achieving precise correlation between the inspection data and the spatial location of the defects at the component level. This lays the foundation for subsequent attribute determination and preliminary scaling assessment. This step integrates the previously scattered inspection images and point cloud data with the structural information of the bridge components, clarifying the spatial location of the defects, the components to which they belong, and their mechanical importance. It solves the problems of ambiguous defect location and disconnection from component structural attributes in traditional methods, providing accurate spatial and structural context for subsequent attribute analysis and scaling determination.

[0043] 2. This invention provides a method for determining disease attributes and initial scaling. Based on bound component information, it extracts all-dimensional attribute parameters of diseases, forming a standardized set of disease attribute parameters. This achieves the transformation of disease attributes from multi-source data to standardized quantification. The process extracts and standardizes parameters from different dimensions for various attributes of diseases, including spatial, geometric, and environmental relationships. It transforms heterogeneous data such as images and point clouds into quantified parameters under a unified standard, eliminating analytical obstacles caused by differences in data formats. As a result, subsequent scaling determination can be based on unified and standardized attribute parameters, ensuring consistency and accuracy in the determination and avoiding attribute analysis biases caused by inconsistent data formats.

[0044] 3. This invention provides a method for determining the attributes of bridge defects and for preliminary scaling assessment. By establishing a structured knowledge base based on industry standards, it automatically matches standardized defect attribute parameter sets with scaling assessment conditions corresponding to the defect types in the knowledge base. This filters out all suitable candidate scaling levels to form a multi-scalar candidate set. Then, based on a scaling selection mechanism for defect nature identification, it conducts a structural safety impact assessment on each scaling in the multi-scalar candidate set. Based on the assessment results, it determines the preliminary scaling and outputs the judgment criteria, thus achieving automated and intelligent scaling assessment from standard matching to safety evaluation. This process transforms expert knowledge from industry standards into structured judgment conditions. Combined with a defect nature identification mechanism, it ensures that scaling assessment conforms to industry standards while fully considering structural safety risks and component importance. Compared to traditional manual assessment, this method significantly improves assessment efficiency and ensures the objectivity and consistency of the results. It solves the problems of subjective interference and low efficiency in manual assessment, providing reliable technical support for the graded management and maintenance decisions of bridge defects. Attached Figure Description

[0045] Figure 1 This is a flowchart of a method for determining disease attributes and performing preliminary scaling according to an embodiment of the present invention;

[0046] Figure 2 This is a schematic diagram of the exposed reinforcement and honeycomb surface of the concrete bottom of the slab according to an embodiment of the present invention.

[0047] Figure 3 This is a schematic diagram illustrating the spalling of the base slab concrete caused by steel reinforcement corrosion, according to an embodiment of the present invention.

[0048] Figure 4 This is a system unit diagram of an embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0050] Example 1

[0051] Please refer to Figure 1 This embodiment 1 provides a method for determining disease attributes and preliminary scaling, including:

[0052] S1. Obtain bridge inspection data, identify the type of defects, bind the defects to specific structural components of the bridge, clarify the spatial distribution boundary of the defects on the component, and record the structural part of the bridge to which the component belongs and its key stress attributes.

[0053] S2. Based on the bound component information, classify and extract the full-dimensional attribute parameters of the disease to form a standardized set of disease attribute parameters;

[0054] S3. Establish a structured knowledge base with industry standards; automatically match the obtained disease attribute parameter set with the scaling judgment conditions of the corresponding disease type in the knowledge base, and filter out all qualified candidate scaling levels to form a multi-scale candidate set;

[0055] S4. Call the preset engineering experience threshold library to conduct structural safety impact assessment on each scale in the multi-scale candidate set; if the assessment confirms that the defect does not affect structural safety, take the lowest value in the candidate scale as the initial assessment scale; if there is a structural safety risk, adjust the scale in combination with the importance level of the component, and finally output the initial assessment result of the defect scale and the judgment basis.

[0056] This embodiment 1 further elaborates on the above steps.

[0057] (1) Component binding and spatial boundary definition

[0058] In bridge defect detection and assessment, accurately locating the defects and determining their spatial distribution is fundamental for subsequent attribute determination and scaling evaluation. First, it is necessary to comprehensively collect 3D point cloud data and high-resolution image data generated during the bridge inspection process. These data provide the 3D spatial geometric information of the bridge structure and the visual information of the defects, respectively. Based on the collected inspection data, please refer to... Figure 2 as well as Figure 3 First, determine the type of damage, including cracks, exposed rebar, wear, deformation, water damage, corrosion, foundation erosion, pavement damage, and other types of damage related to bridge structures, to ensure the comprehensiveness and accuracy of damage classification.

[0059] After identifying the type of defect, a corresponding association is established between the defect and a specific structural component of the bridge. This binds the defect from abstract detection data to concrete physical components, avoiding the problem of ambiguous defect location. To further clarify the specific distribution range of the defect on the component, the defect area needs to be separated from the high-resolution image data using image segmentation algorithms, forming a set containing all pixels in that area. Each pixel's specific location in the image is determined by its horizontal and vertical coordinates. This set comprehensively reflects the location and range of the defect at the image level.

[0060] Since image data only provides two-dimensional information, a mapping relationship between pixel coordinates and three-dimensional point cloud coordinates needs to be established. Using a specialized mapping function, each pixel in the damaged area of ​​the image is converted into a corresponding point in the three-dimensional space of the bridge, forming a three-dimensional point set that corresponds one-to-one with the damaged area in the image. To filter out the points that truly represent the damage characteristics, further calculations are required. Specifically, the method for determining the spatial distribution boundary of the damage on this component is as follows:

[0061] Based on bridge inspection data, three-dimensional spatial data and image data of the components where the defects are located are obtained. The three-dimensional spatial data includes the three-dimensional coordinate information of the components and defects, and the image data is used to locate the two-dimensional area range of the defects. The region feature set of the defects in the image is extracted by the image segmentation method, and the mapping relationship between the region feature set and the three-dimensional spatial data is established, converting the defect region features in the two-dimensional image into the corresponding three-dimensional defect point set.

[0062] The three-dimensional defect point set is optimized to eliminate redundant data, and then a targeted search space is determined based on the geometric features of the three-dimensional defect point set. Candidate component surfaces are retrieved in the search space, and the component surface with the best matching degree is selected as the binding component of the defect by calculating the distance from the representative point in the three-dimensional defect point set to each candidate component surface, thus completing the accurate association between the defect and the component.

[0063] The surface type of the component is identified based on the surface features of the bound component. Then, a local coordinate system is established with the surface of the bound component as the reference. The three-dimensional disease point set is projected onto the two-dimensional plane of the local coordinate system to obtain the two-dimensional disease point set. According to the geometric morphological characteristics of the disease, the spatial distribution boundary of the disease is extracted from the two-dimensional disease point set using the corresponding boundary extraction algorithm. This boundary is used for subsequent calculation of disease attribute parameters and scaling evaluation.

[0064] Specifically, in a preferred embodiment, based on the above process, a specific process for its quantitative implementation is proposed as follows:

[0065] Based on bridge inspection data, 3D point cloud data and high-resolution image data of the components where defects are located are obtained. The 3D point cloud data is represented as follows: ,in For the first in the 3D point cloud data Each element, The three-dimensional spatial coordinates of the corresponding points; simultaneously, the pixel set of the diseased area is extracted using an image segmentation algorithm. , It is a dataset describing the location and extent of diseases in images. For the set of 1 pixel element It is its image pixel coordinates, The horizontal axis is... The vertical axis is... This represents the total number of pixels within the affected area. For traversal index;

[0066] Then, a mapping relationship between pixel coordinates and point cloud coordinates is established. ,in For mapping functions, It is a pixel. The corresponding 3D point cloud coordinates after conversion, subscript Used to associate pixel sets With point cloud collection The elements that form the pixel set. Convert to the corresponding 3D point set ;

[0067] Next, the point set Optimization processing is performed: a point cloud registration algorithm is used to fuse repetitive regions generated by multi-view images, eliminating duplicate point cloud data; then based on... The geometric features automatically determine the search range—area-like diseases are created along the average normal vector direction using the centroid of the point cloud as a reference, with a thickness of [missing information]. Parallel thin-layer search space, The search radius is [value], and its value range is [value]. Linear disease extraction skeleton lines ,by Create a radius of for the axis The cylindrical search space, The range of values ​​is .

[0068] Within the search range, candidate component faces are retrieved using a spatial index. If there is more than one candidate face, then... Sparse sampling generates a representative subset of points ,calculate Distance from each point in the middle to the surface of the candidate component ;

[0069] When the component surface is a plane, the plane equation is used. The distance formula is ,in Let be any point on the reference plane, and let denominate be the magnitude of the plane normal vector;

[0070] When the component surface is a cylindrical surface, the cylinder axis passes through point [point missing]. The direction vector is , radius is The distance formula is Ultimately, the component surface with the smallest average distance and the smallest distance variance was selected as the binding object.

[0071] After completing the component binding, analyze the normal vector of the bound component surface. With global coordinate system The included angle of the axes is used to identify the surface type: It is determined to be the top surface at that time; It is determined to be the bottom surface at that time; The side is determined to be a side view, and the side view is further divided into the left side, right side, front side, or rear side view based on the relationship between the normal vector and the direction of the bridge.

[0072] Finally, boundary delineation is performed: a local coordinate system is established based on the surface of the bound component. The axis is perpendicular to the surface of the component; Projected onto the local coordinate system From a plane, we obtain a two-dimensional point set. ;

[0073] Facial diseases Perform convex hull operations to obtain the convex hull vertex set. , Given the total number of vertices in the convex hull, the vertex set is calculated using the least squares method. By performing polygon fitting, the equation of the closed curve is obtained. ;

[0074] Linear diseases from Extracting skeleton lines Calculate the width of the disease at each point along the skeleton line. Generate a linear disease boundary representation based on skeleton lines and width parameters. This boundary is used for subsequent attribute parameter calculations and scaling evaluation.

[0075] (2) Extraction and standardization of full-dimensional attribute parameters

[0076] After identifying the component to which the disease belongs and its spatial distribution boundaries, in order to achieve systematic analysis of disease attributes and subsequent scaling determination, it is necessary to extract disease attribute parameters from multiple dimensions based on the bound component information and form a unified standard parameter set. This process can transform various characteristics of the disease into quantifiable and comparable standardized data, solving the problems of parameter confusion and lack of unified standards in traditional attribute analysis, and providing a foundation for accurate matching of scaling determination conditions.

[0077] The comprehensive attribute parameters cover spatial attributes, geometric attributes, material attributes, and environmental attributes. The extraction of all types of parameters revolves around the characteristics of the defects and the information of the components. The specific extraction process of its standardized defect attribute parameter set is as follows:

[0078] Based on the defined spatial distribution boundaries of the disease and the component association information, a unified attribute parameter quantification framework is constructed:

[0079] For planar diseases, spatial and geometric attributes are extracted based on the convex hull closed boundary curve of the disease. Spatial attributes include the location determined by the boundary center point; geometric attributes include the disease area calculated using a surface integral algorithm, with the area formula being... ,in This represents the projected region of the boundary onto the surface; it also includes a depth determined by statistical values ​​of the distance from the defect point cloud to the component surface, denoted as... ,in For disease point cloud aggregation, For the surface of the component;

[0080] For linear diseases, spatial and geometric attributes are extracted based on the disease's skeleton line and width distribution. Spatial attributes include the orientation determined by the skeleton line direction, the location determined by the boundary center point, and the distribution pattern determined by the skeleton line length and width distribution. Geometric attributes include the skeleton line length calculated using a line integral algorithm, with the length formula as follows: ,in This is the curve arc length parameter, used for parametric skeleton lines. The skeleton line represents a linear disease; it also includes the overall width calculated based on the skeleton line: first calculate the width of any point on the skeleton line. unit tangent vector at With unit normal vector normal vector Perpendicular to the local direction at that point, along the normal vector Direction determines the boundary points on both sides of the diseased area. and The local width of a point is obtained by calculating the Euclidean distance between the two points. After traversing the skeleton lines, take the maximum value of all local widths as the overall width. The overall width is output as a key geometric parameter; it also includes a depth determined by the maximum distance from the defect point cloud to the component surface, expressed as... ,in The extent of the diseased area;

[0081] When material properties have test data, valid test data are filtered by the distance between the coordinates of the test point and the coordinates of the boundary center point, and the statistical mean is taken as the quantification result: the center point of the spatial distribution boundary of the defect is used as the quantification result. Based on this, data from monitoring points covering the affected area were selected as the valid dataset. These monitoring points included data on concrete strength, carbonation depth, chloride ion content, and resistivity, and their coverage area needed to represent the material condition of the affected area. The selection criteria for the valid dataset were the monitoring points. With the boundary center point The distance is less than the preset threshold The valid dataset is represented as dist ,in The detection value corresponding to the detection point;

[0082] Environmental correlation attributes are obtained through an external data input interface and are used to assess the impact of external conditions on disease development, including environmental corrosivity factors and operational load influence factors. The environmental corrosivity factor can be constructed using a normalized weighted model, with the formula as follows: ,in and These are the measured concentrations of chloride and sulfate ions. and This serves as an industry reference concentration threshold. , The weights are adjustable; the operational load impact factor can be constructed based on the Miner damage principle, and the formula is as follows: ,in For the first Traffic volume of this type of vehicle within the statistical period Average axle load, For standard reference axle load, The fatigue index of the material. For the statistical period, based on the above process, all attribute parameters are standardized into a parameter set according to a preset data format. .

[0083] After extracting all attribute parameters, these parameters are standardized according to the preset data format. Parameters of different dimensions and types are uniformly incorporated into the standardized framework to form a disease attribute parameter set with a clear structure and uniform format, laying the foundation for subsequent matching with the scaling judgment conditions in industry standards.

[0084] (3) Canonical matching and construction of multi-scale candidate sets

[0085] Traditional disease scaling relies on subjective judgment by manual comparison with industry standards. This is not only inefficient but also prone to inconsistencies due to differences in individual understanding. Furthermore, industry standards are mostly textual descriptions, lacking structured judgment logic, making them difficult to use directly for automated analysis. To solve this problem, it is necessary to first establish a structured knowledge base of industry standards, systematically organizing and transforming the scaling judgment conditions corresponding to various diseases into a set of standardized logical expressions that can be recognized and processed by computers.

[0086] This knowledge base covers all common disease types. Each disease type has its own unique logical expression for different scaling levels. Each logical expression contains a logical combination of quantitative and qualitative conditions, clearly defining the judgment criteria for each attribute parameter under that scaling level. After the knowledge base is built, the previously formed standardized disease attribute parameter set is automatically matched with the scaling judgment conditions for the corresponding disease type in the knowledge base.

[0087] During the matching process, the standardized disease attribute parameter set is first established. Extract the quantized values ​​of each attribute parameter. ,in The number of attribute parameters is determined and matched with the scaling conditions for the corresponding disease type in the structured knowledge base. The matching process includes two core steps: quantitative condition matching and qualitative condition matching.

[0088] Quantitative condition matching focuses on the geometric parameters and material properties of the disease, processing and matching quantitative values ​​according to industry-standard calculation methods. This includes area proportion calculation, crack length ratio calculation, and material parameter evaluation. The expression for area proportion calculation is as follows: ,in This represents the percentage of the area affected by disease. The area affected by the disease. The surface area of ​​the component;

[0089] The expression for calculating the crack length ratio is as follows: ,in The ratio of crack lengths. The length of the crack. These are the cross-sectional dimensions of the component;

[0090] Material parameter evaluation requires the inclusion of material property parameters (such as chloride ion content) from the standardized parameter set. resistivity (and the threshold specified by industry-standard criteria) By comparing the results, they can be directly mapped to a quantitative scale.

[0091] After completing the above calculations and comparisons, the quantitative results are matched with the quantitative judgment threshold range of the corresponding disease type in the knowledge base. Each successfully matched quantitative result corresponds to one or more scale level qualitative conditions. The matching focuses on the spatial attributes and appearance characteristics of the disease, and the judgment and matching are completed in accordance with the descriptive standards specified in the industry specifications.

[0092] Based on the spatial attribute information (such as disease direction, distribution pattern, and surface type of the component) and related data of appearance characteristics in the standardized parameter set, and in accordance with the descriptive requirements of the qualitative judgment conditions in the knowledge base, a qualitative judgment result is obtained, which directly corresponds to a specific scale level.

[0093] After matching is completed, all scale levels obtained from quantitative condition matching and qualitative condition four-matching are collected and integrated into the candidate set, ultimately forming a multi-scale candidate set. ,in The total number of candidate scales.

[0094] (4) Safety impact assessment and initial scale determination

[0095] The multi-scale candidate set includes multiple scale levels that meet the judgment criteria of industry standards, but the preliminary assessment results still need to be further confirmed in conjunction with the structural safety impact to avoid ignoring actual safety risks or wasting maintenance resources based solely on standards.

[0096] During the evaluation process, the first step is to select a scale based on the identification of the nature of the disease, targeting multiple candidate scales. This process relies on the key stress attributes of the components and the identification results of the disease type recorded in the component binding process to complete the determination of the nature of the disease.

[0097] When the defect belongs to the superficial defect category, including surface wear, non-structural network cracks, functional layer defects, etc., and the component where the defect is located is a non-critical load-bearing component, a conservative scaling mapping mechanism is activated, and the minimum value in the multi-scale candidate set is selected as the initial evaluation scale; when the defect belongs to the structural defect category, including structural cracks, exposed reinforcement, shear cracks, etc., or the component where the defect is located is a critical load-bearing component, a safety priority scaling selection mechanism is activated, and the maximum value in the multi-scale candidate set is selected as the initial evaluation scale.

[0098] The determination of the nature of the disease needs to be based on the correlation analysis between the key stress attributes of the component and the spatial distribution characteristics of the disease. The key stress components are automatically identified through their structural parts and stress characteristics.

[0099] Finally, the initial assessment results of the disease scaling are output, and a complete set of judgment criteria including the basis for disease nature identification, component location correlation analysis, and scale optimization path are generated;

[0100] At the same time, the system fully records a standardized set of disease attribute parameters, which covers material properties and environmental related attributes, providing a detailed data foundation for subsequent disease development trend judgment and mechanism analysis.

[0101] The assessment method described in Example 1 has broad application prospects in the assessment of crack defects in various reinforced concrete bridges, such as long-span box girder bridges and continuous rigid frame bridges. For bridges with long service lives and whose webs are prone to shear or shrinkage cracks, this method can effectively solve the problems of one-sided parameter extraction and subjective scaling in traditional manual assessments through precise boundary positioning, full-dimensional attribute quantification, and scaling determination that combines standards and experience. It provides traceable and standardized assessment results for bridge maintenance, and is especially suitable for the rapid assessment of bridge defects on busy traffic sections such as highways and national and provincial trunk roads, ensuring the scientific and timely nature of maintenance decisions.

[0102] Furthermore, this method can be extended to the routine monitoring and maintenance management system of bridge defects. By accumulating standardized defect attribute data and scaling assessment results, a database of bridge defect development trends can be formed, providing data support for preventive maintenance. In the maintenance management of large-scale bridge groups, its automated and intelligent assessment process can significantly improve the efficiency of defect assessment and reduce labor costs. At the same time, by balancing the "lowest-cost principle" with safety threshold verification, it can achieve a reasonable allocation of maintenance resources, avoiding waste caused by over-maintenance and preventing the omission of safety risks. This has important practical significance for improving the safety level of bridge structures and extending the service life of bridges.

[0103] Example 2

[0104] Please refer to Figure 4 This embodiment 2 provides a disease attribute determination and initial scaling system, including:

[0105] The component binding and spatial boundary definition unit is used to acquire bridge inspection data, identify the type of defects, bind the defects to specific structural components of the bridge, clarify the spatial distribution boundary of the defects on the component, and record the bridge structural part and key stress attributes to which the component belongs.

[0106] The full-dimensional attribute parameter extraction and standardization unit is used to extract full-dimensional attribute parameters of diseases based on the bound component information, forming a standardized set of disease attribute parameters;

[0107] The standard matching and multi-scale candidate set construction unit is used to establish a structured knowledge base of industry standards; the obtained disease attribute parameter set is automatically matched with the scaling judgment conditions of the corresponding disease type in the knowledge base, and all qualified candidate scaling levels are selected to form a multi-scale candidate set;

[0108] The safety impact assessment and scale preliminary evaluation unit is used to call the preset engineering experience threshold library to conduct structural safety impact assessments on each scale in the multi-scale candidate set. If the assessment confirms that the defects do not affect structural safety, the lowest value among the candidate scales is taken as the preliminary evaluation scale. If there is a structural safety risk, the scale is adjusted in combination with the importance level of the component, and the preliminary evaluation results of the defect scale and the judgment basis are finally output.

[0109] Example 3

[0110] This embodiment 3 also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can implement any step of a disease attribute determination and scaling preliminary evaluation method.

[0111] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0112] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments; further details will not be repeated here.

[0113] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for determining disease attributes and preliminary scaling, characterized in that, include: S1. Obtain bridge inspection data, identify the type of defects, bind the defects to specific structural components of the bridge, clarify the spatial distribution boundary of the defects on the component, and record the structural part of the bridge to which the component belongs and its key stress attributes. S2. Based on the bound component information, classify and extract the full-dimensional attribute parameters of the disease to form a standardized set of disease attribute parameters; S3. Establish a structured knowledge base with industry standards; The obtained disease attribute parameter set is automatically matched with the scaling judgment conditions of the corresponding disease type in the knowledge base, and all qualified candidate scaling levels are selected to form a multi-scale candidate set. S4. Scale selection of multi-scale candidate sets based on disease risk identification; If the assessment confirms that the defects do not affect structural safety, the lowest value among the candidate scales is taken as the initial assessment scale; if there is a structural safety risk, the highest value among the candidate scales is taken as the initial assessment scale. The final output is the initial assessment result of the defect scale and the judgment basis. The method for determining the spatial distribution boundary of the defects on the component is as follows: Based on bridge inspection data, three-dimensional spatial data and image data of the components where the defects are located are obtained. The three-dimensional spatial data includes the three-dimensional coordinate information of the components and defects, and the image data is used to locate the two-dimensional area range of the defects. The region feature set of the defects in the image is extracted by the image segmentation method, and the mapping relationship between the region feature set and the three-dimensional spatial data is established, converting the defect region features in the two-dimensional image into the corresponding three-dimensional defect point set. The three-dimensional defect point set is optimized to eliminate redundant data, and then a targeted search space is determined based on the geometric features of the three-dimensional defect point set. Candidate component surfaces are retrieved in the search space. By calculating the distance from representative points in the three-dimensional defect point set to each candidate component surface, the component surface with the best matching degree is selected as the binding component of the defect, thus completing the accurate association between the defect and the component. The surface type of the component is identified based on the surface features of the bound component. Then, a local coordinate system is established with the surface of the bound component as the reference. The three-dimensional disease point set is projected onto the two-dimensional plane of the local coordinate system to obtain the two-dimensional disease point set. According to the geometric morphological characteristics of the disease, the corresponding boundary extraction algorithm is used to extract the spatial distribution boundary of the disease from the two-dimensional disease point set. This boundary is used for subsequent calculation of disease attribute parameters and scaling evaluation. The specific process for constructing the standardized disease attribute parameter set in S2 is as follows: For planar defects, spatial and geometric attributes are extracted based on the convex hull closed boundary curve of the defect. The spatial attributes include location information determined by the boundary center point; the geometric attributes include the defect area calculated using a surface integral algorithm and the defect depth determined by statistical values ​​of the distance from the defect point cloud to the component surface. For linear defects, spatial and geometric attributes are extracted based on the defect skeleton line and width distribution. The spatial attributes include the orientation determined by the skeleton line direction, the position determined by the boundary center point, and the distribution pattern determined by the skeleton line length and width distribution. The geometric attributes include the skeleton line length calculated using the line integral algorithm, the overall width of the defect obtained based on the skeleton line normal traversal, and the defect depth determined by the maximum distance from the defect point cloud to the component surface. Then, based on the distance between the detection point and the center point of the disease boundary, the valid detection data are screened, and the statistical mean of the valid data is used as the quantitative value of the disease material properties. Then, environmental correlation attributes are obtained through external data interfaces, and environmental corrosion factors are constructed using a normalized weighted model. Operational load impact factors are constructed based on the Miner damage principle. The spatial attributes, geometric attributes, material attributes, and environmental related attributes are organized according to a preset format to form a standardized set of disease attribute parameters.

2. The method for determining disease attributes and preliminary scaling according to claim 1, characterized in that, The types of defects in S1 include: cracks, spalling and exposed reinforcement, wear, deformation, water damage, corrosion, foundation erosion, and pavement damage.

3. The method for determining disease attributes and preliminary scaling according to claim 1, characterized in that, The full-dimensional attribute parameters in S2 include spatial attributes, geometric attributes, material attributes, and environmental attributes.

4. The method for determining disease attributes and preliminary scaling according to claim 1, characterized in that, The knowledge base in S3 contains scaling criteria for various diseases.

5. The method for determining disease attributes and preliminary scaling according to claim 1, characterized in that, The construction process of the multi-scale candidate set in S3 is as follows: Based on a standardized set of disease attribute parameters, quantitative values ​​of each attribute are extracted and matched with scaling conditions for the corresponding disease type in a structured knowledge base. The matching includes quantitative condition matching and qualitative condition matching. Quantitative condition matching takes geometric parameters and material property parameters as the core, calculates the proportion of diseased area and crack length ratio according to industry standards, compares the material property parameters with standard thresholds, and maps the calculation and comparison results to the corresponding scaling level. Qualitative condition matching is based on the spatial attributes and apparent characteristics of the disease, and the corresponding scale level is obtained by comparing it with the qualitative description judgment criteria in the knowledge base. All scale levels obtained from quantitative and qualitative condition matching are integrated to form a multi-scale candidate set.

6. The method for determining disease attributes and preliminary scaling according to claim 1, characterized in that, The confirmation process for the initial assessment results of the disease scale in S4 is as follows: For the multiple candidate scales formed, scale selection is carried out based on the identification of the nature of the disease; based on the key stress attributes of the components and the identification results of the disease type recorded in the component binding process, the nature of the disease is determined. When the disease belongs to the apparent disease category and the component where the disease is located is a non-critical load-bearing component, the conservative scaling mapping mechanism is activated, and the minimum value in the multi-scale candidate set is selected as the initial evaluation scale. When the defect belongs to the category of structural defects, or when the component where the defect is located is a critical load-bearing component, the safety priority scale selection mechanism is activated, and the maximum value in the multi-scale candidate set is selected as the initial evaluation scale. The determination of the nature of the disease needs to be based on the correlation analysis between the key stress attributes of the component and the spatial distribution characteristics of the disease. The key stress components are automatically identified through their structural parts and stress characteristics. Finally, the initial assessment results of the disease scaling are output, and a complete set of judgment criteria including the basis for disease nature identification, component location correlation analysis, and scale optimization path are generated; At the same time, the system fully records a standardized set of disease attribute parameters, which covers material properties and environmental related attributes, providing a detailed data foundation for subsequent disease development trend judgment and mechanism analysis.

7. A disease attribute determination and preliminary scaling system, used to implement the disease attribute determination and preliminary scaling method as described in claim 1, characterized in that, include: The component binding and spatial boundary definition unit is used to acquire bridge inspection data, identify the type of defects, bind the defects to specific structural components of the bridge, clarify the spatial distribution boundary of the defects on the component, and record the bridge structural part and key stress attributes to which the component belongs. The full-dimensional attribute parameter extraction and standardization unit is used to extract full-dimensional attribute parameters of diseases based on the bound component information, forming a standardized set of disease attribute parameters; The normative matching and multi-scale candidate set construction unit is used to build a structured knowledge base of industry standards; The obtained disease attribute parameter set is automatically matched with the scaling judgment conditions of the corresponding disease type in the knowledge base, and all qualified candidate scaling levels are selected to form a multi-scale candidate set. The safety impact assessment and initial scaling determination unit is used to select a scaling factor from a multi-scale candidate set based on disease risk identification. If the assessment confirms that the defects do not affect structural safety, the lowest value among the candidate scales shall be taken as the initial assessment scale. If there is a structural safety risk, the highest value among the candidate scales is taken as the initial assessment scale, and the final output of the initial assessment results of the disease scale and the judgment basis are given.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by a processor as described in any one of claims 1-6: a method for determining disease attributes and performing initial scaling.

Images