Bridge and tunnel maintenance project whole-process intelligent management method and system

Abstract

The invention relates to the technical field of data processing and engineering project process management, in particular to a full-process intelligent management method and system for bridge and tunnel maintenance engineering. The method comprises the following steps: extracting a disease geometric contour based on bridge and tunnel inner wall three-dimensional point cloud data, and calculating a centroid coordinate, a volume and a surface contour area; constructing a space retrieval sphere by using a target disease centroid, screening an adjacent disease set, and calculating a local aggregation density by combining the volume and the centroid distance; calculating the average depth of the target disease, and deducing the overlapping sharing rate of the working plane; constructing a group topology consumption equivalent through linear analysis; and inputting the extreme gradient lifting model to output an optimal repair material dosage table and a material distribution operation instruction book. According to the method, the disease space distribution and aggregation degree can be accurately evaluated, the construction loss surface overlapping probability and the material sharing potential are scientifically evaluated, the material preparation accuracy is improved, waste and cost are reduced, and the whole-process intelligent management of bridge and tunnel maintenance engineering is realized.

Description

Technical Field

[0001] This invention relates to the field of data processing and engineering project process management technology, and in particular to a method and system for intelligent management of the entire process of bridge and tunnel maintenance engineering. Background Technology

[0002] In bridge and tunnel maintenance engineering practice, the efficiency of data flow in the testing and construction stages directly determines the level of project quality control and cost control. Taking the maintenance scenario of mountain highway tunnels as an example, after the testing equipment scans and identifies defects such as lining cracks, the construction unit needs to prepare various repair materials such as epoxy resin in advance based on the defect detection data.

[0003] Traditional material preparation and calculation methods often use the geometric dimensions of a single defect as a basis to calculate the material consumption separately. However, in actual construction, when multiple adjacent defects are treated together, the actual material consumption is often significantly lower than the calculated value. The core reason is that in the repair work of adjacent defects, the construction loss area and the material overflow reserve area overlap spatially, which can realize the sharing and utilization of some repair materials. Traditional calculation methods fail to cover this engineering characteristic.

[0004] Currently, extreme gradient boosting algorithms are commonly used to predict the amount of repair materials needed. These algorithms convert the features of defects such as cracks and spalling collected by equipment like LiDAR into multi-dimensional feature vectors. After model calculation, the algorithm directly outputs a list of repair material categories and quantities. However, in practical applications, traditional extreme gradient boosting algorithms treat each defect as an independent entity and do not take into account the spatial distribution relationships between defects. In the actual environment of bridges and tunnels, defects often exhibit spatial clustering and co-occurring characteristics. For adjacent defects with concentrated distribution, the predicted material consumption values ​​given by traditional extreme gradient boosting algorithms deviate from the actual consumption on site. Over-prepared materials not only cause direct waste of resources but also increase the material management costs and control difficulties at the construction site, thus affecting the implementation and standardized application of a full-process intelligent management system for bridge and tunnel maintenance projects. Summary of the Invention

[0005] To address the problem that traditional extreme gradient boosting algorithms tend to overlook the spatial clustering effect of defects when predicting construction material preparation for bridge and tunnel maintenance projects, leading to the accumulation of material usage and waste, and thus affecting the implementation and standardized application of a full-process intelligent management system for bridge and tunnel maintenance projects, this invention provides a full-process intelligent management method and system for bridge and tunnel maintenance projects.

[0006] Firstly, the present invention provides an intelligent management method for the entire process of bridge and tunnel maintenance engineering, which adopts the following technical solution: A method for intelligent management of the entire process of bridge and tunnel maintenance includes: extracting the geometric contours of physical entities of each defect based on the collected 3D point cloud data of the inner wall of the bridge and tunnel, and calculating the centroid spatial coordinates, volume, and surface contour area of ​​each defect; for target defects among the defects, constructing a spatial retrieval sphere with the centroid of the target defect as the center, filtering out the defect set of all neighboring defects inside the spatial retrieval sphere, and calculating the local cluster density of the target defect by combining the volume of each neighboring defect in the defect set and the distance between the centroid of each neighboring defect and the target defect; calculating the ratio of the volume of the target defect to the surface contour area to obtain the surface area of ​​the target defect. By combining the average depth with the local cluster density, the physical probability of overlapping work loss surfaces of the defect set in actual construction is extrapolated, and the work surface overlap sharing rate of the target defect is calculated. Linear analysis is performed through the work surface overlap sharing rate to determine the weight allocation of the spatial characteristics of the defect set on the target defect, and a group topological consumption equivalent reflecting the actual degree of material loss is constructed. The group topological consumption equivalent is input into an extreme gradient boosting model that has been trained with historical topological relationship data in advance for regression calculation, outputting an optimal repair material usage table, and generating a material distribution operation guide for on-site construction, realizing intelligent management of the entire process of bridge and tunnel maintenance engineering.

[0007] This invention assesses the spatial distribution characteristics of defects by constructing a spatial retrieval sphere and calculating local clustering density, accurately reflecting the degree of defect clustering and providing a reliable spatial distribution basis for subsequent calculation of work surface overlap and sharing rate. By calculating the work surface overlap and sharing rate, it scientifically assesses the probability of overlapping construction loss surfaces, comprehensively considering the influence of average defect depth and local clustering density, accurately reflecting the material sharing potential of adjacent defects in actual construction. Furthermore, by constructing a group topological consumption equivalent, it comprehensively assesses the actual degree of material loss, accurately reflecting the impact of spatial correlation between defects on material consumption, and providing feature inputs containing spatial topological information for extreme gradient boosting models. Material consumption prediction based on population topology consumption equivalent improves the accuracy of repair material preparation, effectively reduces material waste and on-site management costs, and realizes intelligent management of the entire process of bridge and tunnel maintenance engineering.

[0008] Furthermore, the local cluster density satisfies: In the formula, Target disease Local aggregation density, Target disease The collection of diseases, For adjacent diseases in a disease cluster volume, For adjacent diseases in a disease cluster With the target disease The Euclidean distance between the centroids, This is the standard arm span length for manual labor. It is a natural exponential function.

[0009] This invention achieves a scientific assessment of local cluster density by constructing a spherical density model that includes volume weights and distance attenuation terms. It more accurately reflects the density of diseases within the working range with a standard manual working arm span as the radius. The distance attenuation term ensures that the contribution of neighboring diseases to the cluster density decreases with increasing distance, thereby effectively assessing the spatial clustering characteristics of diseases.

[0010] Furthermore, the overlap and sharing rate of the work surfaces satisfies: In the formula, Target disease The overlap and sharing rate of the work surface Target disease Local aggregation density, Target disease The standard construction tolerance cross-sectional area for the applicable construction process To cause the target disease Reference cross-sectional area at that time Target disease average depth, To cause the target disease Reference depth at time It is a natural exponential function.

[0011] This invention achieves a scientific assessment of the overlap and sharing rate of the work surface by constructing an exponential decay model that includes local cluster density and geometric ratio. It more accurately reflects the impact of the degree of disease clustering on the overlap of the construction tolerance area. The geometric ratio term corrects the effect of different disease types and depths on the overlap probability, thereby effectively assessing the degree of material sharing between adjacent diseases in actual construction.

[0012] Furthermore, the population topology consumption equivalent satisfies: In the formula, Target disease The population topology consumption equivalent, Target disease volume, Target disease The overlap and sharing rate of the work surface Target disease The collection of diseases, For adjacent diseases in a disease cluster The volume.

[0013] This invention achieves a scientific assessment of the equivalent topological consumption of a disease population by constructing a product model that includes correction terms for disease volume and overlap sharing rate. This model more accurately reflects the actual material consumption after considering the spatial overlap effect. The correction term reflects the moderating effect of the volume ratio among diseases on the sharing effect, thereby effectively characterizing the actual topological consumption characteristics of disease groups and providing accurate feature input for predicting material consumption.

[0014] Furthermore, after constructing the group topology consumption equivalent that reflects the actual degree of material loss, the method further includes: in response to the fact that the value of the group topology consumption equivalent is less than the preset consumption lower limit of the target disease, assigning the group topology consumption equivalent to the preset consumption lower limit.

[0015] This invention achieves reasonable assurance of the consumption equivalent of the group topology by setting a preset consumption lower limit and applying numerical constraints. It ensures that even in the case of high overlap and sharing, the necessary minimum material consumption is still maintained, preventing material shortage problems caused by over-correction and improving the reliability and security of material consumption prediction.

[0016] Furthermore, the calculation of the centroid spatial coordinates, volume, and surface contour area of ​​each disease includes: extracting the geometric contour of each disease physical entity using a point cloud region growing algorithm; calculating the centroid spatial coordinates and volume of each disease based on the point cloud bounding box; and calculating the surface contour area of ​​each disease based on the polygon area of ​​the point cloud contour.

[0017] Furthermore, the training logic of the extreme gradient boosting model, which has been pre-trained with historical data on topological relationships, includes: extracting historical construction records, calling a point cloud region growing algorithm to parse archived historical point cloud data with a point density not less than a preset density threshold; reconstructing the topological relationships of historical defects and selecting a set of historical neighboring defects, and then calculating the historical group topological consumption equivalent corresponding to the historical defects; associating the historical group topological consumption equivalent with historical actual material consumption records to generate a supervised learning sample set to complete the training of the extreme gradient boosting model.

[0018] Furthermore, when constructing the spatial retrieval sphere, the radius of the spatial retrieval sphere adopts the standard human arm span length, and the value range of the standard human arm span length is as follows: rice.

[0019] Furthermore, when the reference cross-sectional area and reference depth that match the target disease cannot be obtained, the reference cross-sectional area and reference depth are set to preset values, and a system log alarm is triggered.

[0020] Secondly, this invention provides an intelligent management system for the entire process of bridge and tunnel maintenance engineering, which adopts the following technical solution: A smart management system for the entire process of bridge and tunnel maintenance engineering includes a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the aforementioned smart management method for the entire process of bridge and tunnel maintenance engineering is realized.

[0021] By adopting the above technical solution, a computer program is generated from the above-mentioned intelligent management method for the entire process of bridge and tunnel maintenance engineering, and stored in the memory so that it can be loaded and executed by the processor. In this way, a terminal device can be made based on the memory and the processor for convenient use.

[0022] The present invention has the following technical effects: (1) In view of the problem that traditional material preparation calculation ignores the engineering characteristics of overlapping loss areas and shared use of materials when dealing with adjacent diseases in a concentrated manner, resulting in a significant deviation between the calculated value and the actual consumption, this invention constructs a spatial retrieval sphere to screen adjacent diseases, calculates the local cluster density to quantify the spatial clustering degree of diseases, and combines the average depth of diseases to extrapolate the probability of overlapping loss surfaces in the operation, accurately calculates the overlap and sharing rate of the operation surface, effectively breaking the limitation of traditional independent calculation of a single disease, and incorporating the material sharing and loss overlap characteristics during centralized treatment into the calculation system, thereby reducing the deviation between the predicted value of material usage and the actual consumption on site, and avoiding direct material waste caused by over-preparation.

[0023] (2) In view of the shortcomings of traditional extreme gradient boosting algorithms, which assume that the defects are independent individuals and do not include spatial relationships, this invention completes the weight allocation by the overlap and sharing rate of the working surface, and constructs a group topological consumption equivalent that reflects the actual degree of material loss. It transforms the core engineering factors such as spatial distribution relationship and associated characteristics between defects into indicators that the model can identify. At the same time, the model completes training based on historical data of topological relationships and can learn the actual material consumption pattern under different spatial distributions. The output of the optimal repair material usage table is more in line with the actual construction of bridge and tunnel maintenance site, effectively solving the problem of large prediction deviation of traditional models and providing accurate and reliable data support for construction material preparation.

[0024] (3) By accurately predicting the amount of materials needed, we can avoid the waste of materials caused by over-preparation of materials, and reduce the management costs and manpower input of excess materials in the storage, transfer and control of the construction site. At the same time, we can avoid the problems of construction interruption and delay due to insufficient material preparation, and make the material preparation work of the construction unit more targeted and reasonable.

[0025] (4) This invention realizes the full-process digital and automated flow from the acquisition of three-dimensional point cloud data of bridge and tunnel inner walls and the automated extraction of physical characteristics of defects, to spatial correlation analysis, accurate prediction of material usage and automatic generation of work instructions. It effectively solves the problems of low data flow efficiency, excessive manual intervention and non-standard process in the detection and construction links of traditional bridge and tunnel maintenance projects, and provides technical support for the standardized implementation of the intelligent management system for the whole process of bridge and tunnel maintenance. Attached Figure Description

[0026] Figure 1 This is a flowchart of a method for intelligent management of the entire process of bridge and tunnel maintenance engineering according to an embodiment of the present invention.

[0027] Figure 2 This is a schematic diagram comparing the material usage of a single defect dimension in a smart management method for the entire process of bridge and tunnel maintenance engineering according to an embodiment of the present invention.

[0028] Figure 3 This is a schematic diagram comparing the total usage of various material categories in a smart management method for the entire process of bridge and tunnel maintenance engineering according to an embodiment of the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] This invention discloses an intelligent management method for the entire process of bridge and tunnel maintenance engineering, referring to... Figure 1 This includes steps S001-S005: S001: Based on the collected 3D point cloud data of the inner wall of the bridge and tunnel, extract the geometric contour of each physical entity of the disease, and calculate the centroid spatial coordinates, volume and surface contour area of ​​each disease.

[0031] Specifically, the three-dimensional LiDAR on the inspection vehicle is used to scan and obtain three-dimensional point cloud data of the inner wall of the bridge and tunnel. The coordinate system is metric. The geometric contours of each physical entity of the disease are extracted using the point cloud region growth algorithm. The centroid spatial coordinates and volume of each disease are calculated based on the point cloud bounding box. The surface contour area of ​​each disease is calculated based on the polygon area of ​​the point cloud contour.

[0032] S002: For the target disease among all diseases, construct a spatial retrieval sphere with the centroid of the target disease as the center, filter out the disease set of all neighboring diseases inside the spatial retrieval sphere, and calculate the local cluster density of the target disease by combining the volume of each neighboring disease in the disease set and the distance between the centroid of each neighboring disease and the target disease.

[0033] It should be noted that by constructing a spatial retrieval sphere with the standard manual operation arm span as its radius, the actual construction work area can be accurately matched, incorporating the actual operational characteristics of the construction scenario into the assessment of the degree of defect clustering. Therefore, the local clustering density calculated in this step can accurately reflect the spatial clustering state of defects in actual construction, providing a basis that closely aligns with actual construction practices for calculating the overlap and sharing rate of the work surface.

[0034] Specifically, when constructing the spatial retrieval sphere, the radius of the sphere adopts the standard human working arm span. The human working arm span typically refers to the radius that a person can comfortably reach with a handheld tool, such as a spray gun or grouting pipe. In the confined space of a tunnel, this value is generally taken as 0.8 meters. In this embodiment, the value range is [range missing]. Meters, capable of covering the entire population for work.

[0035] Specifically, the local cluster density satisfies: ; In the formula, Target disease Local aggregation density, Target disease The collection of diseases, For adjacent diseases in a disease cluster volume, For adjacent diseases in a disease cluster With the target disease The Euclidean distance between the centroids, This is the standard arm span length for manual labor. It is a natural exponential function.

[0036] Among them, with the volume of adjacent diseases Increased distance from adjacent diseases The shrinkage of the molecular attenuation summation value significantly increases, leading to a final dimensionless localized density of the disease. The nonlinear increase reflects that the cluster of diseases in the working space where the target disease is located is more dense, and the potential for mutual interference between the construction surfaces is greater.

[0037] S003: Calculate the ratio of the volume of the target defect to the surface contour area to obtain the average depth of the target defect. Based on the average depth, use the local cluster density to extrapolate the physical probability of the overlap of the operation loss surface of the defect set in actual construction, and calculate the operation surface overlap sharing rate of the target defect.

[0038] It should be noted that the overlap and sharing rate of work surfaces is a core indicator for measuring the degree of material sharing during the construction of adjacent defects. The average depth of the defects directly affects the amount of material filled and the extent of overflow during construction. The deeper the defect, the greater the depth requirement for material filling, the less material overflows onto the work surfaces of adjacent defects, and the lower the probability of overlap and sharing. Therefore, the overlap and sharing rate of work surfaces calculated in this step can accurately assess the potential for material sharing during the construction of adjacent defects, providing a key basis for the accurate calculation of material usage.

[0039] Specifically, the overlap and sharing rate of the work surfaces satisfies: ; In the formula, Target disease The overlap and sharing rate of the work surface Target disease Local aggregation density, Target disease The standard construction tolerance cross-sectional area for the relevant construction process is set based on the physical influence range of the actual construction process. For example, for lining cracks, considering the need to leave edge overlap space when manually or mechanically grooving and applying fabric, the standard construction tolerance cross-sectional area is set on both sides of the crack. m represents the operational tolerance, and the corresponding standard construction tolerance cross-sectional area is taken as [value missing]. m²; To cause the target disease Reference cross-sectional area at that time Target disease average depth, To cause the target disease Reference depth at time The reference cross-sectional area and reference depth are natural exponential functions. When the reference cross-sectional area and reference depth for matching the target defect cannot be obtained, the reference cross-sectional area and reference depth are set to preset values, and a system log alarm is triggered. In this embodiment, the system presets the reference cross-sectional area to 0.5m² and the reference depth to 0.2m. The above preset values ​​are determined based on the statistical median of historical tunnel maintenance cases, which can cover most common shallow defects, such as local hollowing or small-area spalling, ensuring that the automated system can still generate basic instructions when data is missing, thus avoiding calculation interruption.

[0040] Among them, the local cluster density When the standard construction tolerance characteristics increase, the absolute value of the negative values ​​in the index term increases, and the overlap and sharing rate of the work surface increases. Approaching the upper limit of 1 indicates that the material overflow areas of dense defects highly overlap under conventional tolerance construction. At the same time, when the average depth of the target defect itself increases, the target defect's ability to absorb material in depth becomes stronger, overflow interference decreases, and the overlap sharing rate decreases.

[0041] S004: By performing linear analysis on the overlap and sharing rate of the working surface, the weight allocation of the spatial characteristics of the disease set on the target disease is determined, and a group topological consumption equivalent reflecting the actual degree of material loss is constructed.

[0042] It should be noted that the group topology consumption equivalent is the key to transforming the spatial correlation of defects into a material consumption indicator. It comprehensively considers the volume of a single defect and the operational overlap effect between multiple defects, thus better reflecting the material consumption patterns in actual construction. Therefore, this step uses the overlap and sharing rate of the work surface for weight allocation, incorporating the spatial correlation and volume ratio between defects into the assessment of material loss, and comprehensively considering the impact of the overall characteristics of defect clusters on material consumption.

[0043] Specifically, the population topology consumption equivalent satisfies: ; In the formula, Target disease The population topology consumption equivalent, Target disease volume, Target disease The overlap and sharing rate of the work surface Target disease The collection of diseases, For adjacent diseases in a disease cluster The volume.

[0044] Specifically, when the volume of the target disease occupies a relatively small portion of the disease cluster area, the subsequent difference term approaches 1. At this point, the target disease greatly benefits from the cluster protection effect, and its consumption equivalent characteristic... A significant downward reduction occurred.

[0045] Specifically, after constructing the population topological consumption equivalent that reflects the actual degree of material loss, the following is also included: To prevent insufficient construction materials due to excessively low calculated values, a preset lower limit for consumption is forcibly set, i.e., when... hour, , It is determined based on experience in minimizing construction filling losses for minor defects. It can take into account both the filling needs of minor defects and the rationality of construction material preparation, so as to ensure that the construction losses of the attached minor cracks do not need to be calculated separately, but the gaps themselves still need to be filled in a small amount.

[0046] S005: Input the equivalent of group topology consumption into the extreme gradient boosting model that has been trained with historical data of topological relationships in advance for regression calculation, output the optimal repair material usage table, and generate a material distribution operation instruction for on-site construction, so as to realize the intelligent management of the whole process of bridge and tunnel maintenance projects.

[0047] Specifically, the training logic of the extreme gradient boosting model, which has been pre-trained using historical data on topological relationships, includes: Historical construction records are extracted, and a point cloud region growth algorithm is called to parse the archived historical point cloud data with a point density not less than a preset density threshold. In this embodiment, the preset density threshold is 100 points / m², which is an empirical value for parsing bridge and tunnel defect detection point cloud data. This density can ensure that the point cloud data can clearly identify the defect outline and details, meet the accuracy requirements of historical defect topology reconstruction, and avoid excessive computation due to excessive point cloud density. The topological relationships of historical diseases are reconstructed and the set of historical neighboring diseases is selected, and then the topological consumption equivalent of the historical group corresponding to the historical disease is calculated. By associating historical population topological consumption equivalents with historical actual material consumption records, a supervised learning sample set is generated to complete the training of the extreme gradient boosting model.

[0048] like Figure 3 As shown in the figure, the horizontal axis represents the disease number to be treated, and the vertical axis represents the amount of repair material used for a single disease. The figure shows the material usage calculation results of the traditional method and the present invention for each disease. It can be seen from the figure that the calculated usage of the present invention method is lower than that of the traditional method for all diseases to be treated. This effectively eliminates the material redundancy caused by the traditional method ignoring the spatial aggregation effect of diseases, and achieves precise control of material usage at the level of a single disease.

[0049] like Figure 3 As shown in the figure, the total usage calculation results of the traditional method and the present invention are compared for four types of commonly used repair materials in bridge and tunnel maintenance projects: pressure grouting, epoxy resin grouting, polymer repair mortar, and waterproof coating. It can be seen from the figure that the method of the present invention can achieve a certain amount of savings in usage for all types of commonly used repair materials. This verifies that by incorporating the spatial topological association features of the defects and constructing a group topological consumption equivalent optimization model input, the present invention can effectively reduce the deviation between the predicted value of material usage and the actual consumption on site, reduce material waste and engineering costs, and realize intelligent management of the entire process of bridge and tunnel maintenance projects.

[0050] This invention also discloses an intelligent management system for the entire process of bridge and tunnel maintenance engineering, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the intelligent management method for the entire process of bridge and tunnel maintenance engineering according to this invention is realized.

[0051] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0052] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for intelligent management of the entire process of bridge and tunnel maintenance engineering, characterized in that, include: Based on the collected 3D point cloud data of the inner wall of the bridge and tunnel, the geometric contours of each physical entity of the disease are extracted, and the centroid spatial coordinates, volume and surface contour area of ​​each disease are calculated. For each disease, a spatial retrieval sphere is constructed with the centroid of the target disease as the center. The disease set of all neighboring diseases inside the spatial retrieval sphere is selected. The local clustering density of the target disease is calculated by combining the volume of each neighboring disease in the disease set and the distance between the centroid of each neighboring disease and the target disease. Calculate the ratio of the volume of the target defect to the surface contour area to obtain the average depth of the target defect. Combine the average depth with the local cluster density to deduce the physical probability of the overlap of the working loss surface of the defect set in actual construction, and calculate the working surface overlap sharing rate of the target defect. By performing linear analysis on the overlap and sharing rate of the work surface, the weight allocation of the spatial characteristics of the disease set on the target disease is determined, and a group topological consumption equivalent reflecting the actual degree of material loss is constructed. The equivalent consumption of the group topology is input into an extreme gradient boosting model that has been trained with historical data of topological relationships for regression calculation. The optimal repair material consumption table is output, and a material distribution operation instruction for on-site construction is generated, realizing intelligent management of the entire process of bridge and tunnel maintenance projects.

2. The intelligent management method for the entire process of bridge and tunnel maintenance engineering according to claim 1, characterized in that, The local cluster density satisfies: ； In the formula, Target disease Local aggregation density, Target disease The collection of diseases, For adjacent diseases in a disease cluster volume, For adjacent diseases in a disease cluster With the target disease The Euclidean distance between the centroids, This is the standard arm span length for manual labor. It is a natural exponential function.

3. The intelligent management method for the entire process of bridge and tunnel maintenance engineering according to claim 1, characterized in that, The overlap and sharing rate of the work surfaces satisfies: ； In the formula, Target disease The overlap and sharing rate of the work surface Target disease Local aggregation density, Target disease The standard construction tolerance cross-sectional area for the applicable construction process To cause the target disease Reference cross-sectional area at that time Target disease average depth, To cause the target disease Reference depth at time It is a natural exponential function.

4. The intelligent management method for the entire process of bridge and tunnel maintenance engineering according to claim 1, characterized in that, The population topology consumption equivalent satisfies: ； In the formula, Target disease The population topology consumption equivalent, Target disease volume, Target disease The overlap and sharing rate of the work surface Target disease The collection of diseases, For adjacent diseases in a disease cluster The volume.

5. The intelligent management method for the entire process of bridge and tunnel maintenance engineering according to claim 1, characterized in that, After constructing the population topological consumption equivalent that reflects the actual degree of material loss, the following is also included: In response to the fact that the value of the population topology consumption equivalent is less than the preset consumption lower limit value of the target disease, the population topology consumption equivalent is assigned the value of the preset consumption lower limit value.

6. The intelligent management method for the entire process of bridge and tunnel maintenance engineering according to claim 1, characterized in that, The calculation of the centroid spatial coordinates, volume, and surface contour area of ​​each disease includes: The geometric contours of each diseased physical entity are extracted using a point cloud region growing algorithm; The centroid spatial coordinates and volume of each disease are calculated based on the point cloud bounding box. The surface contour area of ​​each disease is calculated based on the polygon area of ​​the point cloud contour.

7. The intelligent management method for the entire process of bridge and tunnel maintenance engineering according to claim 1, characterized in that, The training logic for an extreme gradient boosting model pre-trained on historical topological data includes: Extract historical construction records and use a point cloud region growth algorithm to parse archived historical point cloud data with point density not less than a preset density threshold; The topological relationships of historical diseases are reconstructed and the set of historical neighboring diseases is selected, and then the topological consumption equivalent of the historical group corresponding to the historical disease is calculated. By associating historical population topological consumption equivalents with historical actual material consumption records, a supervised learning sample set is generated to complete the training of the extreme gradient boosting model.

8. The intelligent management method for the entire process of bridge and tunnel maintenance engineering according to claim 2, characterized in that, When constructing the spatial retrieval sphere, the radius of the spatial retrieval sphere adopts the standard human arm span length, and the value range of the standard human arm span length is as follows: rice.

9. The intelligent management method for the entire process of bridge and tunnel maintenance engineering according to claim 3, characterized in that, When the reference cross-sectional area and reference depth that match the target disease cannot be obtained, the reference cross-sectional area and reference depth are set to preset values, and a system log alarm is triggered.

10. A smart management system for the entire process of bridge and tunnel maintenance engineering, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method for intelligent management of the entire process of bridge and tunnel maintenance engineering as described in any one of claims 1-9 is implemented.

Images