Building structure crack self-diagnosis-self-repair system and method

Abstract

The invention discloses a building structure crack self-diagnosis-self-repair system and method, and relates to the technical field of building engineering data processing, and the method comprises the steps: synchronously collecting the resistance time sequence data, strain space distribution data and environmental erosion parameters of a crack through an embedded sensor array when the crack appears in a building structure; calculating the variable quantity of the resistance time sequence data in unit time as an expansion rate value, scanning the strain space distribution data at the same time, and identifying the coordinate of the target monitoring point of which the strain value change rate exceeds a critical point; clustering the target monitoring points of which the Euclidean distances are smaller than a preset aggregation threshold value to form a continuous area based on the identified coordinates of the target monitoring points; according to the accumulated duration after the crack appears, dynamic calculation weights are given to the expansion rate value, the point density value and the environmental erosion parameter respectively, and the three values are weighted and summed to determine a comprehensive risk value; and controlling the repair equipment to execute repair operation according to the comprehensive risk value.

Description

Technical Field

[0001] This application relates to the field of building engineering data processing technology, and more specifically, to a self-diagnosis and self-repair system and method for building structural cracks. Background Technology

[0002] Throughout the entire lifecycle of a building structure, the formation and development of cracks are common defects affecting structural safety, durability, and functionality. Concrete and steel structures are highly susceptible to cracks of varying degrees under the influence of loads, environmental erosion, and material aging. If not detected and addressed promptly, these cracks can continue to propagate, potentially leading to serious safety incidents such as structural instability. Therefore, efficient self-diagnosis of building structural cracks, enabling real-time perception and assessment of crack conditions, is crucial for ensuring the safe operation of building structures and reducing maintenance costs.

[0003] While existing technologies for self-diagnosis and repair of structural cracks have achieved some degree of automated monitoring, significant shortcomings remain. For example, some technologies fail to adequately consider all factors influencing crack development during the diagnosis process, often relying solely on a single physical parameter, which is insufficient to fully reflect the actual development of the cracks. Other technologies employ fixed evaluation criteria in the risk assessment phase, failing to dynamically adjust the assessment logic according to different stages of crack development. This can easily lead to biased risk assessments, thereby affecting the targetedness and timeliness of repair measures.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a self-diagnosis and self-repair system and method for building structure cracks to solve the above-mentioned technical problems.

[0006] This application provides a self-diagnosis and self-repair system for building structural cracks, including:

[0007] The crack data acquisition module is used to synchronously acquire the resistance time series data, strain spatial distribution data and environmental erosion parameters of the cracks through a pre-embedded sensor array when cracks appear in the building structure.

[0008] The target monitoring point identification module is used to calculate the change in the resistance time series data per unit time as the expansion rate value, and simultaneously scan the strain spatial distribution data to identify the coordinates of the target monitoring points where the strain value change rate exceeds the critical point; wherein, the strain value change rate is the absolute value of the strain difference between adjacent monitoring points divided by the distance between monitoring points.

[0009] The point density value determination module is used to cluster target monitoring points whose Euclidean distance between them is less than a preset aggregation threshold into continuous regions based on the coordinates of the identified target monitoring points; and to calculate the ratio of the number of target monitoring points included in each continuous region to the area of ​​the circumscribed convex hull of the region as the point density value of the region.

[0010] The risk determination module is used to dynamically calculate weights for the propagation rate value, the point density value, and the environmental erosion parameter based on the cumulative time after the crack appears, and then sum the three weighted values ​​to determine the comprehensive risk value.

[0011] The repair module is used to control the repair equipment to perform repair operations based on the comprehensive risk value.

[0012] This application provides a self-diagnosis and self-repair method for building structural cracks, including:

[0013] When cracks appear in the building structure, the resistance time series data, strain spatial distribution data and environmental erosion parameters of the cracks are collected synchronously through a pre-embedded sensor array.

[0014] The change in the resistance time series data per unit time is calculated as the expansion rate value. At the same time, the strain spatial distribution data is scanned to identify the coordinates of the target monitoring points where the strain value change rate exceeds the critical point. The strain value change rate is the absolute value of the strain difference between adjacent monitoring points divided by the distance between the monitoring points.

[0015] Based on the coordinates of the identified target monitoring points, target monitoring points whose Euclidean distance to each other is less than a preset aggregation threshold are clustered into continuous regions; the ratio of the number of target monitoring points in each continuous region to the area of ​​the circumscribed convex hull of the region is calculated as the point density value of the region.

[0016] Based on the cumulative duration after the crack appears, the propagation rate value, the point density value, and the environmental erosion parameter are dynamically weighted and then weighted and summed to determine the comprehensive risk value.

[0017] The repair equipment is controlled to perform repair operations based on the comprehensive risk value.

[0018] Based on the embodiments provided in this application, by synchronously collecting the resistance time series data, strain spatial distribution data and environmental erosion parameters of cracks through a pre-embedded sensor array, it is possible to simultaneously obtain multi-dimensional information reflecting crack development, avoiding the limitations of relying on a single physical parameter for judgment, thereby more comprehensively capturing the characteristics and influencing factors of cracks and improving the completeness and accuracy of self-diagnosis; at the same time, by performing cluster analysis on the target monitoring points and calculating the point density value, the spatial distribution range and concentration of cracks can be accurately identified, providing basic data that is more in line with the actual crack morphology for subsequent risk assessment.

[0019] By dynamically assigning different weights to the propagation rate, point density, and environmental erosion parameters based on the cumulative duration after crack appearance, the calculation of the comprehensive risk value can be adapted to the characteristics of cracks at different development stages. This overcomes the risk judgment bias caused by using fixed evaluation standards, making the risk assessment results more consistent with the actual development trend of cracks. The comprehensive risk value obtained by combining the above multi-dimensional data and dynamic weights can more reliably guide the operation of repair equipment, ensuring the pertinence and timeliness of repair measures, and effectively guaranteeing the safe operation of building structures. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0021] Figure 1 This is a structural diagram of an optional self-diagnosis and self-repair system for building structural cracks according to an embodiment of this application;

[0022] Figure 2 This is a flowchart of an optional self-diagnosis and self-repair method for building structure cracks according to an embodiment of this application;

[0023] Figure 3 This is a flowchart of another optional self-diagnosis-self-repair method for building structure cracks according to an embodiment of this application;

[0024] Figure 4 This is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application.

[0025] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0026] 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 a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0027] According to one aspect of the embodiments of this application, a self-diagnosis and self-repair system for building structural cracks is also provided. For example... Figure 1 As shown, the system includes:

[0028] The crack data acquisition module 101 is used to synchronously acquire the resistance time series data, strain spatial distribution data and environmental erosion parameters of the crack through a pre-embedded sensor array when cracks appear in the building structure.

[0029] The target monitoring point identification module 102 is used to calculate the change in resistance time series data per unit time as the expansion rate value, and at the same time scan the strain spatial distribution data to identify the coordinates of the target monitoring points in which the strain value change rate exceeds the critical point; wherein, the strain value change rate is the absolute value of the strain difference between adjacent monitoring points divided by the distance between monitoring points.

[0030] The point density value determination module 103 is used to cluster target monitoring points whose Euclidean distance between them is less than a preset aggregation threshold into continuous regions based on the coordinates of the identified target monitoring points; and to calculate the ratio of the number of target monitoring points included in each continuous region to the area of ​​the circumscribed convex hull of the region as the point density value of the region.

[0031] The risk determination module 104 is used to dynamically calculate the weights of the propagation rate value, the point density value, and the environmental erosion parameter based on the cumulative time after the crack appears, and then sum the three weighted values ​​to determine the comprehensive risk value.

[0032] Repair module 105 is used to control the repair equipment to perform repair operations based on the comprehensive risk value.

[0033] According to another aspect of the embodiments of this application, such as Figure 2 As shown, this application provides a self-diagnosis and self-repair method for building structural cracks, including:

[0034] S201, When cracks appear in the building structure, the resistance time series data, strain spatial distribution data and environmental erosion parameters of the cracks are collected synchronously through a pre-embedded sensor array;

[0035] In some embodiments, a pre-embedded sensor array refers to a sensor network pre-embedded within a building structure (such as a concrete beam or steel structure node), including resistance sensors, strain gauges, and environmental erosion sensors (such as temperature, humidity, and chloride ion concentration sensors), forming a distributed monitoring grid. Its function is to synchronously capture multi-dimensional data during crack propagation; for example, a strain gauge array pre-embedded in the tension zone of a concrete beam can monitor changes in the strain field around the crack in real time.

[0036] For example, inside a bridge box girder, sensors are arranged in a matrix with a spacing of 10cm×10cm. When a crack passes through a certain area, the surrounding sensors simultaneously collect data on resistance jump (caused by the crack interrupting the conductive path), strain change, and environmental humidity, providing basic data for subsequent analysis.

[0037] In S201, a pre-embedded sensor array synchronously collects resistance time-series data, strain spatial distribution data, and environmental erosion parameters, overcoming the limitations of single-parameter monitoring. Resistance data reflects changes in crack conductivity, strain data reflects structural mechanical response, and environmental parameters relate to external influencing factors. These three elements work together to form a multi-dimensional diagnostic matrix, comprehensively capturing crack characteristics from electrical, mechanical, and environmental perspectives. This provides three-dimensional data support for subsequent analysis and avoids misjudgments caused by incomplete information.

[0038] S202, calculate the change in resistance time series data per unit time as the expansion rate value, and at the same time scan the strain spatial distribution data to identify the coordinates of the target monitoring point where the strain value change rate exceeds the critical point; wherein, the strain value change rate is the absolute value of the strain difference between adjacent monitoring points divided by the distance between monitoring points.

[0039] The critical point refers to the threshold of the strain rate of change, used to determine the critical state of crack propagation. This value is set based on the mechanical properties of the structural materials (such as the tensile strength of concrete) and historical monitoring data, and is usually determined through finite element simulation or laboratory testing. For example, for concrete beams, the critical point is set to 0.002 (absolute strain difference / spacing). If the spacing between adjacent monitoring points is 10 cm, when the strain difference exceeds 0.0002 (i.e., 0.0002 / 0.1 = 0.002), it is determined that there is a risk of crack propagation in that area. In steel structure joints, the critical point may be set to 0.0015, corresponding to the upper limit of the strain rate of change before the steel yields.

[0040] It should be explained that the strain change rate is used to quantify the degree of drastic change in strain in space. This indicator reflects the stress concentration at the crack tip, and the larger the value, the more significant the crack propagation trend.

[0041] In S202, the dynamic changes in the time dimension are combined with the critical state in the spatial dimension. The propagation rate quantifies the speed of crack development, and the strain value change rate accurately locates the critical point of stress concentration. This captures both the temporal evolution trend of the crack and the spatial risk monitoring points, laying the point-based foundation for subsequent spatial analysis. In S203, based on the coordinates of the identified target monitoring points, target monitoring points with a mutual Euclidean distance less than a preset aggregation threshold are clustered into continuous regions. The ratio of the number of target monitoring points in each continuous region to the area of ​​the circumscribed convex hull of that region is calculated as the point density value of that region.

[0042] The preset aggregation threshold is used to cluster the Euclidean distance threshold of target monitoring points. When the distance between two points is less than this value, they are determined to belong to the same continuous crack area. The value needs to be determined in conjunction with the structural type and crack propagation characteristics. For example, in concrete floor slabs, the preset aggregation threshold is set to 5cm to ensure that scattered micro-crack points are aggregated into continuous areas. In steel structure weld monitoring, the threshold is set to 2cm because steel crack propagation is more concentrated, requiring more refined clustering.

[0043] The circumscribed convex hull area of ​​the region contains the area of ​​the smallest convex polygon of all target monitoring points, used to quantify the spatial distribution range of the crack region. For example, when three monitoring points form a triangle, the circumscribed convex hull area is the area of ​​that triangle. This indicator, combined with the point density value (number of points / area), can assess the density of cracks. For example, if the circumscribed convex hull area of ​​a certain region is 200 cm²... 2 If there are 10 monitoring points, the point density is 0.05 points / cm². 2 This indicates that the cracks are sparsely distributed; if the area is 50cm² 2 If there are 12 points, the point density value is 0.24 points / cm². 2 It requires intensive repair.

[0044] In S203, discrete monitoring points are transformed into structured spatial region features. Clustering operations identify the continuous distribution range of cracks through preset aggregation thresholds, the area of ​​the circumscribed convex hull quantifies the physical scale of the region, and the point density intuitively reflects the spatial concentration of cracks. This transforms abstract monitoring data into spatial morphological indicators that can be directly used for risk assessment, improving the spatial resolution of diagnosis.

[0045] S204. Based on the cumulative time since the crack appeared, the propagation rate value, point density value, and environmental erosion parameter are dynamically weighted and the three are weighted and summed to determine the comprehensive risk value.

[0046] In S204, risk assessment is adapted to the characteristics of the entire crack life cycle. The initial crack propagation rate has a greater impact, while the effects of environmental erosion and spatial distribution become more prominent in the middle and later stages. Dynamic weights, adjusted over time, allow the comprehensive risk value to match the dominant factors at different stages in real time, avoiding the assessment distortion caused by fixed weights and making risk judgments more in line with actual development trends.

[0047] S205, control the repair equipment to perform repair operations based on the comprehensive risk value.

[0048] In S205, the comprehensive risk value integrates multi-dimensional data and dynamic weights as the core basis for repair decisions, ensuring that repair operations are neither over-interventionist (avoiding resource waste) nor delayed in response (preventing risk expansion), so that repair measures are accurately matched with the actual degree of damage of the cracks, thereby improving the pertinence and economy of repair.

[0049] Furthermore, such as Figure 3 As shown, the repair operations performed by the equipment controlled according to the comprehensive risk value include:

[0050] S301, determine the crack risk level by matching the comprehensive risk value with the preset risk level table, sort the point density values ​​from largest to smallest, and select the continuous area with the first preset proportion as the high stress concentration area.

[0051] In some embodiments, the preset risk level table is a grading system based on comprehensive risk values, typically including three levels: low, medium, and high, corresponding to different repair strategies. This table needs to be developed in conjunction with structural design specifications and historical repair data. For example, a comprehensive risk value of 0-30 indicates low risk, 30-60 indicates medium risk, and above 60 indicates high risk. For instance, if a crack has a comprehensive risk value of 45, it is determined to be medium risk according to the table and requires repair using a medium-viscosity repair material (e.g., 800-1200 cP) and fiber reinforcement density of 150% of the base density.

[0052] The preset proportions include, but are not limited to, the top 20% or 30%. This proportion needs to be adjusted according to structural safety requirements. For high-risk structures (such as nuclear power plant buildings), it may be set to the top 10%.

[0053] In S301, risk level classification establishes a repair priority framework, while high-stress zone marking identifies the most hazardous areas through point density sorting, shifting repair operations from "comprehensive coverage" to "key breakthroughs," improving resource utilization efficiency while ensuring safety and avoiding ineffective investment in low-risk areas.

[0054] S302, based on the crack risk level, query the preset viscosity mapping table to determine the target viscosity range of the repair material, and map the coordinate range of the high stress concentration area to the fiber density reinforced area.

[0055] In some embodiments, the preset viscosity mapping table is a lookup table that associates risk levels with the viscosity range of repair materials. For example, low risk corresponds to 500-800 cP, medium risk to 800-1200 cP, and high risk to 1200-1500 cP. Viscosity selection needs to balance material flowability and bond strength; high-viscosity materials are suitable for filling wide cracks, while low-viscosity materials are used for penetrating micro-cracks. For example, if a crack has a high risk level, the target viscosity range of 1200-1500 cP is determined by referring to the table, and an epoxy resin-based high-viscosity material is selected. If the ambient humidity is high, the upper limit of the viscosity range may be adjusted to 1600 cP to prevent premature curing of the material.

[0056] In S302, the properties of the repair material are dynamically adapted to the characteristics of the crack risk. The viscosity range is adjusted according to the risk level (e.g., high-viscosity materials are used to enhance the bond in high-risk areas), and the fiber density reinforcement zone corresponds to the stress concentration area, so that the physical properties of the material are specifically matched with the structural stress requirements.

[0057] S303 controls the repair equipment to perform repair operations, including: preparing the repair material matrix according to the target viscosity range of the repair material; spraying fiber-reinforced repair material at a fiber density higher than the base density in the fiber density-reinforced area; and calculating the spatial gradient vector of strain data in the high stress concentration area in real time, and controlling the fiber nozzle to spray in the opposite direction of the strain gradient.

[0058] In S303, material properties are combined with construction techniques to enhance the mechanical effectiveness of the repair. Adapting material viscosity to risk levels ensures construction feasibility and curing effectiveness; high fiber density reinforcement zones directly counteract stress concentration; and reverse gradient spraying creates a crack-resistant barrier between the fiber direction and the stress transmission path. These three elements synergistically improve the overall integrity and crack resistance of the repaired structure, achieving a coupling of materials, processes, and stress.

[0059] The base density refers to the reference spray density of the fiber-reinforced repair material, which is typically set based on the material's mechanical properties and structural design requirements. For example, the base density for ordinary concrete repair is 100 g / m³. 2 This applies to general cracks; for steel structure repair, a setting of 150g / m² may be used. 2 To enhance fatigue resistance.

[0060] For example, the basic density is set to 120 g / m³. 2 In the fiber density enhancement area, spray at 1.5 times, i.e., 180g / m². 2 This increases the shear strength of the crack. If the crack width exceeds 2 mm, the foundation density may need to be increased to 150 g / m³. 2 The spray ratio was increased to 2 times (300g / m³). 2 ).

[0061] It should be explained that the spatial gradient vector is a vector describing the direction and rate of strain change in space. Its modulus reflects the severity of strain change, and its direction points towards the direction of the fastest strain increase. This vector is obtained by calculating the spatial distribution of strain differences between adjacent monitoring points and is used to guide the fiber injection direction. For example, in a high-stress area, the spatial gradient vector points northeast with a modulus of 0.008 / cm, indicating that the crack propagation trend is northeast. Reverse gradient injection involves controlling the nozzle to move southwest, making the fiber perpendicular to the crack propagation direction, thus enhancing the crack-stopping effect.

[0062] The repair material is sprayed in the opposite direction to the spatial gradient vector, i.e., towards the region of least strain change. This inverse strain gradient spraying strategy allows fibers to form a cross network at the crack tip, effectively preventing further crack propagation. For example, if the spatial gradient vector is 30° east of north, the nozzle sprays in a direction 30° southwest of south, while adjusting the spraying speed according to the gradient modulus length; the larger the modulus length, the slower the speed, ensuring that the material fully fills the crack.

[0063] Based on the embodiments provided in this application, the crack risk level is determined by matching the comprehensive risk value with a preset risk level table, and high stress concentration areas are marked, so that the repair operation can be precisely focused on the areas where the crack hazard is more prominent; the target viscosity range of the repair material is determined by combining the risk level, and the high stress concentration areas are mapped to fiber density reinforcement areas, so that the physical properties of the repair material are adapted to the crack risk level, and the fiber reinforcement measures are more in line with the actual needs of stress concentration; controlling the repair equipment to spray fibers in the direction of the reverse strain gradient can make the fiber distribution direction match the stress transmission path of the structure, improve the cooperative stress-bearing capacity of the repair material and the original structure, and enhance the overall integrity of the repaired structure.

[0064] Furthermore, after controlling the repair equipment to perform repair operations based on the comprehensive risk value, the method also includes:

[0065] Monitor the recovery of resistance in the repaired area. If the amount of resistance recovery per unit time is lower than the recovery threshold set according to the crack risk level during the diagnostic stage, increase the flow rate of the curing agent.

[0066] The recovery threshold refers to the standard for the recovery of resistance value per unit time based on the crack risk level during the diagnostic phase. The higher the risk level, the stricter the threshold (requiring faster recovery), used to determine whether the curing effect meets the standard during the repair process. For example, the recovery threshold for low-risk cracks (comprehensive risk value 0-30) is set at 5Ω / h (resistance value recovers at least 5Ω per hour); for high-risk cracks (above 60), it is set at 15Ω / h, because high-risk cracks require faster curing to prevent propagation.

[0067] In some embodiments, increasing the curing agent flow rate refers to increasing the mixing ratio of the curing agent and the repair material (achieved by increasing the flow rate), thereby accelerating the curing reaction of the material and preventing secondary crack propagation due to slow curing. For example, when the original curing agent flow rate is 10 ml / min, the resistance recovery is 3 Ω / h (below the threshold of 5 Ω / h). After increasing the flow rate to 15 ml / min, the curing speed is accelerated, and the resistance recovery increases to 6 Ω / h, meeting the repair requirements.

[0068] After the repair material has cured, a standard test load is applied to the cured area. The maximum fluctuation of the resistance value under the load and the reciprocal of the time required for the strain value to recover to the initial value as a set percentage after unloading are collected as the strain recovery rate.

[0069] In some embodiments, applying standard test loads refers to loads that simulate the normal service state of the structure (such as uniformly distributed loads and concentrated loads in design specifications) to verify the recovery of the bearing capacity of the repaired area.

[0070] For example, a 2kN / m² pressure is applied to the repaired concrete slab. 2 By simulating the weight of personnel and equipment under a uniformly distributed load, and by monitoring resistance fluctuations and strain recovery, the cooperative stress-bearing capacity of the repair material and the original structure is evaluated.

[0071] The initial value percentage refers to the proportion of the strain value after repair that recovers to the initial strain value before the crack appeared. It reflects the degree of recovery of the structural mechanical properties, and the value must be determined in conjunction with the structural functional requirements (such as higher requirements for load-bearing structures). For example, the initial value is set to 90% for key load-bearing structures such as bridge main beams (the strain needs to recover to 90% of the initial value); the initial value is set to 70% for non-load-bearing walls, allowing for a certain amount of residual strain.

[0072] Based on the maximum fluctuation range of the resistance value and the strain recovery rate, the weight distribution ratio of the propagation rate value, point density value, and environmental erosion parameters is readjusted, and the preset risk level table and preset viscosity mapping table are updated.

[0073] In some embodiments, updating the preset risk level table and preset viscosity mapping table refers to adjusting the level classification threshold and viscosity range based on the resistance fluctuation and strain recovery data after repair, so that the table better reflects the actual repair effect. For example, if the strain recovery rate is generally low after repair of the original medium-risk level (30-60), the upper limit of medium risk is lowered to 50, and the corresponding viscosity range is adjusted from 800-1200 cP to 1000-1400 cP to enhance the material adhesion.

[0074] Based on the embodiments provided in this application, monitoring the recovery of the resistance value in the repair area and dynamically adjusting the flow rate of the curing agent can optimize the curing effect according to the real-time status of the repair process, avoiding the impact of insufficient or excessive curing on the repair quality. After the repair is completed, the maximum fluctuation range of resistance and the strain recovery rate are obtained through standard test loads, providing feedback based on the actual repair effect for weight adjustment. Based on the above feedback, the weight allocation ratio is readjusted and the preset table is updated, so that subsequent risk assessment and repair strategies can continuously adapt to the actual state of the structure.

[0075] Furthermore, based on the cumulative duration after crack appearance, dynamic calculation weights are assigned to the propagation rate value, point density value, and environmental erosion parameters, including:

[0076] A time-dependent function for crack development stages is constructed, mapping the cumulative duration to stage coefficients. The weight of the expansion rate value decreases with the increase of the stage coefficient according to a hyperbolic tangent curve.

[0077] In some embodiments, the stage coefficient is a quantitative indicator of cumulative duration (such as a value between 0 and 1), reflecting whether the crack is in the early stage (small coefficient), middle stage (medium coefficient), or late stage (large coefficient), providing a time dimension basis for dynamic weighting. For example, a cumulative duration of 1 day (early stage) maps to a coefficient of 0.2, 10 days (middle stage) maps to 0.6, and 30 days (late stage) maps to 0.9, allowing the weight to transition naturally with the development stage of the crack.

[0078] The hyperbolic tangent curve characteristic is a slow decay in the early stage and a rapid decay in the later stage, which conforms to the crack development law. That is, the initial propagation rate has a large impact on risk (high weight), and the impact weakens in the later stage as it tends to stabilize (low weight). For example, when the stage coefficient is 0.2 (early stage), the propagation rate has a weight of 0.6; when the coefficient is 0.8 (later stage), the weight drops to 0.2, allowing environmental erosion and spatial distribution factors to dominate the risk assessment.

[0079] For the point density value, calculate the spatial topological connectivity between each continuous region, including: if the Euclidean distance between the boundary points of two regions is less than the preset connectivity threshold, then establish a connection edge, and use the ratio of the number of connection edges to the total number of regions as the regional network density;

[0080] The preset connectivity threshold is a distance threshold between boundary points used to determine whether two consecutive regions are associated, and is used to identify the overall distribution network of cracks (such as whether a through crack has formed). For example, it is set to 8cm for concrete structures (allowing for larger spacing between associated cracks, as cracks tend to extend along the aggregate interface); and 3cm for welded seams in steel structures (requiring closer association, as steel cracks mostly extend linearly).

[0081] It should be explained that boundary points are the edge points of the circumscribed convex hull of a region. A distance less than a threshold indicates that the regions are spatially close (possibly belonging to the same crack system), and establishing connecting edges can reflect the topological correlation of the cracks. For example, if the boundary point of region A is 5cm away from the boundary point of region B (less than the threshold of 8cm), establishing a connecting edge indicates that the two regions may be branches of the same crack, and the overall risk needs to be assessed.

[0082] The ratio of the number of connecting edges to the total number of regions reflects the connectivity of the cracked areas. A higher ratio indicates a more concentrated distribution of cracks (such as forming a network) and a higher risk. For example, if 10 regions form 15 connecting edges, the network density is 15 / 10 = 1.5, indicating strong crack correlation; if 5 regions have only 2 connecting edges, the density is 0.4, indicating that the cracks are dispersed and the risk is lower.

[0083] Multiply the point density value by the natural logarithm of the regional network density to generate a topology-weighted density index as the weight input;

[0084] Among these methods, the natural logarithm processing makes the influence of network density nonlinear (avoiding the dominance of extreme values), and the topology-weighted density reflects both the density of points and regional correlations, better aligning with the actual risks of crack spatial distribution. For example, a point density of 0.24 points / cm². 2 The network density is 1.5 (ln1.5≈0.405), and the index is 0.24×0.405≈0.097, which reflects both the density of points and the superimposed effect of regional connections on risks.

[0085] For environmental erosion parameters, the absolute value is integrated using a sliding time window, and then convolved with the stage coefficient to generate the cumulative erosion intensity.

[0086] Among them, the sliding time window integral (such as the 7-day window) reflects the recent cumulative erosion effect (rather than the instantaneous value), and the convolution with the stage coefficient allows the erosion impact to change dynamically with the crack stage (such as later cracks being more susceptible to long-term erosion). For example, the humidity is high during the rainy season, the 7-day window integral value is large, and after convolution with the mid-term stage coefficient (0.6), the erosion accumulation intensity increases, the weight increases, which is consistent with the characteristic that cracks are more sensitive to the environment in the mid-term.

[0087] Dynamic weights are generated based on a combination of hyperbolic tangent attenuation weights controlled by stage coefficients, topological weighted density indices, and cumulative erosion intensity.

[0088] In one specific implementation, dynamic weights are generated based on the following formula:

[0089]

[0090] Among them, W total The comprehensive dynamic weight is (normalized, 0 to 1); α is the basic coefficient for the propagation rate weight (0.3 to 0.5, 0.4 for concrete); k is the aging attenuation coefficient (0.02 to 0.05 / day, 0.05 for critical structures); t is the cumulative duration of cracks (days). β is the normalized expansion rate (actual rate / material limiting rate, 0 to 1); β is the basic coefficient for point density weighting (0.2 to 0.4, 0.4 for dense regions); D is the normalized point density (0 to 1); net γ is the regional network density (number of connected edges / total number of regions, 0 to 1); γ is the basic coefficient for environmental erosion weight (0.2 to 0.3, 0.3 for high humidity); The normalized erosion integral (sliding window integral / critical threshold, 0 to 1); t stage This is the stage coefficient (cumulative duration mapping value, from 0 to 1, calculated via t / (t+30)).

[0091] It should be noted that the attenuation weight (time dimension), topological weighted density (spatial dimension), and erosion accumulation intensity (environmental dimension) of the stage coefficient control work together to make the weight allocation adapt to the temporal evolution, spatial distribution and external influences of the cracks, and achieve a multi-dimensional dynamic balance.

[0092] Based on the embodiments provided in this application, a time-effect function is constructed to map the cumulative duration to stage coefficients, so that the weight of the propagation rate value dynamically decays with the crack development stage, adapting to the changes in the dominant risk factors of cracks at different stages; by calculating the topological connectivity of regional space and generating a topological weighted density index, the weight of the point density value can reflect the correlation between crack regions, which is more in line with the overall characteristics of crack spatial distribution; by using sliding time window integration and stage coefficient convolution for environmental erosion parameters, the cumulative effect of erosion and its coupling relationship with crack development stages can be reflected, so that the allocation of dynamic weights is more in line with the actual mechanism of multi-factor coupling of cracks.

[0093] Furthermore, controlling the fiber nozzle to spray in the opposite strain gradient direction includes:

[0094] A Delaunay triangular control grid is constructed using the geometric centroids of each continuous region within the high stress concentration zone as control nodes.

[0095] In this context, the geometric centroid is the central coordinate of the region. The Delaunay triangular mesh ensures that the triangular elements are as equilateral as possible (avoiding elongated triangles), uniformly covering the high-stress area and providing stable spatial elements for strain gradient calculation. For example, the centroid coordinates of the three high-stress areas form a triangular mesh, covering the crack tip and a 20cm radius around it, ensuring accurate strain gradient calculation within each triangular element.

[0096] The strain gradient vector at the vertex of each triangular element is calculated in real time, and the maximum value of the vector magnitude within the element is taken as the reference direction.

[0097] In this model, the vertices are the three corner points of the triangular mesh. The strain gradient vector is calculated using the strain values ​​at the vertices (such as the rate of strain change along the x and y directions), reflecting the direction and rate of strain change from high to low within the element. For example, if the strain at vertex A is 300 με, B is 200 με, and C is 100 με, the calculated gradient vector points in the A→C direction, indicating that the strain decreases from A to C at a rate of 200 με / cm.

[0098] In some embodiments, the vector with the largest modulus is the direction of the most drastic strain change (such as the crack propagation direction). The reference direction is used to ensure that the repair spraying direction is specifically designed to combat the high-strain zone. For example, if the modulus of the three vectors within the unit are 200, 150, and 100 με / cm, the direction corresponding to 200 με / cm is taken as the reference direction to guide the nozzle to prioritize combating the strain in that direction.

[0099] The difference between the median of the target viscosity range of the repair material and the current viscosity is mapped to a directional perturbation angle using a hyperbolic tangent function.

[0100] The viscosity difference (target viscosity vs. current viscosity) is converted into a perturbation angle (small angular deviation) using a hyperbolic tangent function. A larger difference results in a larger perturbation angle (suitable for materials with poor flowability), while a smaller difference results in a smaller perturbation angle. For example, a viscosity difference of 800 cP (hyperbolic tangent ≈ 0.76) maps to a perturbation angle of 7.6° (magnified 10 times); a difference of 200 cP (hyperbolic tangent ≈ 0.197) maps to 1.97°, balancing spray flexibility.

[0101] The theoretical injection direction is generated by superimposing the disturbance angle in the opposite direction of the reference direction, while the second spatial derivative of the strain energy density in the theoretical direction projection region is scanned.

[0102] The theoretical direction projection area is the projection range of the theoretical injection direction onto the structural surface (such as the area covered by the nozzle injection path), used to detect the bending characteristics of strain energy within this area (whether there is abrupt curvature change). For example, if the theoretical injection direction is 45° southwest, the projection area is a 10cm × 10cm range covered by this direction. The second derivative of the strain energy density within this area is scanned to determine whether bending cracks exist.

[0103] If the second spatial derivative exceeds the curvature threshold, the injection angle is deflected along the normal of the strain energy density contour line, and the amount of deflection is proportional to the rate of curvature change.

[0104] The second spatial derivative reflects the curvature change of the strain energy (e.g., whether the crack bends). Exceeding the curvature threshold indicates significant bending, requiring adjustment of the injection direction to conform to the crack morphology. For example, the curvature threshold can be set to 1×10⁻⁶. -4 (Concrete), when the second derivative is 1.2 × 10⁻⁶ -4 When this occurs, it indicates that the crack has obvious curvature, and the spray angle needs to be adjusted.

[0105] It should be noted that strain energy density contour lines are curves formed by points with equal strain energy density. The normal direction is perpendicular to the contour line (pointing towards the direction of the fastest change in strain energy). Deflecting along this direction allows the repair material to fill the inner side of the curved crack (the stress concentration point). The strain energy contour lines are arc-shaped with the normal direction being radial. After the nozzle is deflected radially, the material fits more closely to the curved part of the crack, avoiding repair dead zones.

[0106] A large rate of curvature change (steeply curved crack) results in a large deflection, while a small rate of curvature change (gently curved crack) results in a small deflection, ensuring that the deflection amplitude matches the crack morphology.

[0107] In one implementation, the correction expression for the injection angle is:

[0108] θ jet =-θ base +Δθ-λ×sgn(C curv )×|C curv | μ

[0109] Where, θ jet The final fiber spraying angle (°, with the surface normal as 0°); θ base Δθ is the strain gradient reference direction angle (°, direction of maximum vector magnitude); Δθ is the viscosity perturbation angle (°, calculated by 5×tanh(Δη / 500), where Δη is the viscosity difference); λ is the curvature correction coefficient (°·(m)). 4 / J)^μ, concrete is taken as 3.2, steel structure as 4.5), ^ is the symbol for exponentiation; C curv The second spatial derivative of strain energy density (J / m) 4 , reflecting the degree of crack curvature; μ is the curvature sensitivity index (1.1 to 1.3, 1.3 at the crack tip); sgn(C curv ) is a sign function, used to determine the direction of crack bending. Specifically, when C curv When >0, sgn(C curv ) = 1 indicates that the crack bends in a certain direction (e.g., to the right), and the jetting angle deflects in the opposite direction; when C curv When <0, sgn(C) curv ) = -1 indicates that the crack bends in the opposite direction (e.g., to the left), and the jetting angle deflects in the corresponding opposite direction; when C curv When = 0, sgn(C curv ) = 0, indicating that the crack has no bend and no deflection; sgn(C curv The purpose of this is to ensure that the deflection direction of the spray angle matches the bending direction of the crack, thus preventing the repair material from accumulating on the inside or outside of the bend.

[0110] Based on the embodiments provided in this application, a Delaunay triangular control mesh is constructed using the geometric centroid of the high stress concentration area to provide a structured spatial unit division for strain gradient calculation, making the determination of the injection direction more consistent with the stress distribution characteristics of the local area; the reference direction is determined based on the magnitude of the strain gradient vector and the direction perturbation angle is superimposed, which can adapt to the subtle morphological changes of the crack boundary while ensuring the targeting of the injection direction; the injection angle is adjusted by the second spatial derivative of the strain energy density, which allows the fiber injection path to actively adapt to the curvature characteristics of the crack region, avoiding the occurrence of repair weak points at stress abrupt changes.

[0111] Furthermore, the weighting ratio is readjusted based on the maximum fluctuation range of the resistance value and the strain recovery rate, including:

[0112] The maximum fluctuation range of the resistance value is converted into the structural stiffness reduction factor, and the reciprocal of the strain recovery rate is converted into the material relaxation time constant.

[0113] Through a pre-defined calibration mechanism, electrical and mechanical monitoring data are transformed into mechanical parameters describing structural performance. Specifically, based on experimental data from similar structures, a correlation is established between the maximum fluctuation amplitude of resistance and the degree of structural stiffness loss (the greater the fluctuation, the more significant the stiffness reduction). The reciprocal of the strain recovery rate is defined as the time index for the material to recover its elasticity from deformation (the slower the rate, the longer the recovery time). The core of this transformation is to utilize the intrinsic relationship between the change in resistance after crack repair reflecting the degree of crack closure and the strain recovery reflecting the material's elasticity, enabling the monitoring data to directly serve the assessment of the structural mechanical performance.

[0114] A five-dimensional state space is established, consisting of the expansion rate, point density, environmental erosion parameters, structural stiffness reduction coefficient, and material relaxation time constant. The dominant feature plane is extracted through recursive orthogonal decomposition.

[0115] The five-dimensional state space is an analytical system that integrates three core parameters affecting crack risk (spread rate, point density, and environmental erosion) with two mechanical indicators reflecting the repair effect (stiffness reduction factor and relaxation time constant). Recursive orthogonal decomposition iterates through multiple iterations, gradually eliminating parameter combinations with less impact on the repair effect, ultimately retaining the two parameters with the most significant impact on structural performance to form the dominant characteristic plane. This aims to simplify the complexity of multi-parameter coupling, allowing subsequent weight adjustments to focus on key factors. A weight correction vector is constructed within the dominant characteristic plane, including: the spread rate weight correction term, which is proportional to the projection of the structural stiffness reduction factor onto the basis vector of the dominant characteristic plane.

[0116] The basis vectors of the dominant characteristic plane are the direction vectors describing the core associations of that plane. The projection of the stiffness reduction coefficient onto the basis vectors reflects the strength of the association between the spread rate and stiffness loss. The correction term is proportional to this projection, meaning that the more significant the effect of the spread rate on stiffness loss (the larger the projection), the greater its weight will be, thus enabling the dynamic weights to truly reflect the actual impact of the parameters on structural safety.

[0117] The point density weight correction term implements a region fusion mechanism, including: when the difference in relaxation time constants between two consecutive regions is less than the fusion tolerance, merging the regions and recalculating the circumscribed convex hull area and topology weighted density index;

[0118] The fusion tolerance is a threshold used to determine whether the material recovery capabilities of two regions are similar. When the difference in relaxation time constants between two regions is within the tolerance range, it indicates that their material elastic recovery characteristics are similar (they may belong to the same crack system). By merging regions and recalculating spatial parameters (such as circumscribed convex hull and topological density), the analytical bias caused by region segmentation can be eliminated, and the overall distribution characteristics of the cracks can be more accurately reflected. For example, if the fusion tolerance is set to 0.3h, the relaxation time constant of region A is 2.1h, and that of region B is 2.3h (difference 0.2h < 0.3h), then the two regions are merged, and the area of ​​the merged circumscribed convex hull (the sum of the areas of the original two regions minus the overlapping part) and the topological weighted density are recalculated, so that the point density weight can reflect the overall density of the cracks.

[0119] The environmental erosion weight correction term introduces an erosion-stiffness coupling factor, which is the absolute value of the partial derivative of the structural stiffness reduction coefficient with respect to the cumulative erosion intensity.

[0120] Among them, the erosion-stiffness coupling factor is used to quantify the direct impact of environmental erosion on structural stiffness. By analyzing the change (partial derivative) of the structural stiffness reduction coefficient for each unit change in cumulative erosion intensity, the impact of environmental factors on structural safety is transformed into a quantifiable indicator, enabling the weight adjustment of environmental erosion parameters to accurately match their actual impact (the greater the impact, the higher the weight).

[0121] The comprehensive risk value is recalculated using the updated continuous regions and corrected weights, and the preset aggregation threshold and preset connectivity threshold are optimized simultaneously to minimize the variance of the number of merged regions.

[0122] Among these measures, a feedback mechanism is used to optimize the region division parameters. After updating the region division and weights, the dispersion (variance) of the number of merged regions is calculated. By adjusting the aggregation threshold (distance standard for cluster monitoring points) and connectivity threshold (distance standard for judging the correlation between regions), the distribution of the number of regions is made more stable (with minimal variance), ensuring that the region division is neither too fragmented (difficult to assess as a whole) nor excessively merged (masking local risks).

[0123] In one implementation, the expression for the weight correction coefficient is:

[0124]

[0125] Where ΔW is the total weight correction coefficient (from 0 to 0.3); δ r K is the extension rate correction factor (0.1 to 0.2, 0.2 when stiffness loss is large); stiff This is the structural stiffness reduction factor (0 to 1, converted from the maximum resistance fluctuation range); For K stiff The projection (0 to 1) onto the unit basis vectors of the dominant characteristic plane; δ represents the unit basis vectors of the dominant characteristic planes extracted from the five-dimensional state space through recursive orthogonal decomposition, pointing in the direction of the parameter combination that has the greatest impact on the structural stiffness; ρ Point density correction factor (0.15 to 0.25, 0.25 for dense regions); F merge The regional fusion correction coefficient (Nmerge / Ntotal, where Nmerge is the logarithm of the fused regions and Ntotal is the total number of continuous regions within the current monitoring area) is used to normalize the regional fusion correction term and avoid correction bias caused by differences in the number of regions; δ e Environmental erosion correction factor (0.05 to 0.15, 0.15 for highly erosive environments); The erosion-stiffness coupling factor is 0 to 1. The ratio (ranging from 0 to 1) of the sliding time window integral value of environmental erosion parameters (such as the cumulative integral of chloride ion concentration and humidity) to the critical erosion threshold of this type of structure is used to quantify the cumulative degree of environmental erosion (0 indicates no erosion, and 1 indicates that the critical danger value has been reached).

[0126] Based on the embodiments provided in this application, the maximum resistance fluctuation amplitude and strain recovery rate are converted into structural stiffness reduction coefficient and material relaxation time constant, providing a quantitative basis for weight adjustment based on structural mechanical performance. A five-dimensional state space is constructed and the dominant feature plane is extracted through recursive orthogonal decomposition, which can focus on key influencing factors in complex relationships of multi-parameter coupling, making weight correction more targeted. Specific weight correction mechanisms are designed for different parameters (such as the projection correlation between expansion rate weight and stiffness reduction coefficient, and the regional fusion mechanism of point density weight, etc.), which can enable weight adjustment to accurately match the intrinsic relationship between each parameter and structural performance, and improve the dynamic adaptability of risk assessment.

[0127] In one approach, when simultaneously optimizing preset aggregation thresholds and preset connectivity thresholds, regional network co-evolutionary control is executed:

[0128] The continuous regions retained after load testing are grouped according to the relaxation time constant; the standard deviation of the minimum distance between the centroids of each region within each group is calculated as the spatial discretization index; the distribution of the connection edge length defined by weight 3 within each group is statistically analyzed, and the proportion of length values ​​falling into the high-frequency range is used as the connection stability index; when the spatial discretization index exceeds the density tolerance: based on the deviation direction of the current spatial discretization index from the historical mean, the preset aggregation threshold is adjusted in reverse compensation; when the connection stability index is lower than the reliability threshold: the length value with the highest frequency of occurrence is extracted from the historical connection edge length dataset and set as the new preset connectivity threshold; after each round of adjustment, the region fusion operation is re-executed until the double convergence condition is met: the spatial discretization index of all groups is reduced to less than 50% of the initial value; the improvement of the connection stability index for three consecutive iterations is less than the reliability gain threshold.

[0129] Furthermore, when calculating the ratio of the number of target monitoring points included in each continuous region to the area of ​​the circumscribed convex hull of that region, a region integrity correction is performed, including:

[0130] Detect blank sub-regions without target monitoring points inside the circumscribed convex hull, and calculate the proportion of the area of ​​the blank sub-region to the total area of ​​the convex hull as the blank ratio;

[0131] Specifically, using image recognition technology, the interior of the circumscribed convex hull (the smallest convex polygon containing all monitoring points) of a continuous region is scanned. Blank areas not covered by monitoring points are marked, and their area is calculated as a percentage of the total area of ​​the convex hull. The purpose is to determine the completeness of the monitoring point distribution. If the percentage of blank areas is too high, it indicates that the monitoring data may not accurately reflect the regional characteristics and requires further correction.

[0132] Extract the target monitoring point that is closest to the boundary of the blank sub-region and calculate its physical Euclidean distance to the geometric center of the blank sub-region;

[0133] Specifically, by locating the nearest monitoring point at the boundary of the blank sub-region, the straight-line distance between that point and the center of the blank region is calculated to determine the correlation between the blank region and the valid monitoring data. The closer the distance, the greater the influence of the monitoring point data on the blank region (relatively high data reliability); the farther the distance, the worse the representativeness of the data for the blank region.

[0134] When the blank area exceeds the area tolerance and the physical Euclidean distance exceeds the displacement tolerance, the area of ​​the blank sub-region is subtracted from the total area of ​​the convex hull as the effective calculated area.

[0135] The calculation includes setting area tolerance (maximum allowable blank area percentage) and displacement tolerance (maximum allowable distance). When both exceed these limits, it indicates that the blank area significantly interferes with density calculation. The blank area must be removed, and the remaining effective area should be used to calculate the point density to avoid underestimating the point density due to "virtual area" (e.g., a large blank area but few monitoring points would result in a higher actual density). For example, if the area tolerance is set at 15% and the displacement tolerance at 5cm, and the blank area percentage is 20% (exceeding the tolerance) and the nearest point distance is 7cm (exceeding the tolerance), then the effective calculated area = the total convex hull area 200cm². 2 - Blank area 40cm 2 =160cm 2 This ensures that the point density calculation is based on the actual area covered by the monitoring points.

[0136] The corrected point density value is obtained by dividing the number of target monitoring points by the effective calculated area, and the continuous area that triggers the correction is recorded and marked as the broken area.

[0137] The corrected point density value is obtained by dividing the number of target monitoring points by the effective calculated area (the area after deducting blanks), making it more consistent with the actual distribution density of monitoring points. The area that triggers the correction is marked as a "fragmented area" to distinguish areas where the data reliability is low due to the scattered distribution of monitoring points.

[0138] When assigning dynamic weights to the point density values ​​in the subsequent process, an inverse proportional attenuation factor based on displacement tolerance is applied to the corrected point density values ​​of the broken regions.

[0139] For fractured areas, an attenuation factor is set proportionally to the displacement tolerance based on the actual distance between the nearest monitoring point and the center of the blank area (the factor decreases with greater distance), attenuating the correction point density value. The aim is to reduce the weight of areas with low data reliability in risk assessment and avoid misjudgments caused by scattered monitoring points. For example, if the displacement tolerance is 5cm and the nearest point in the fractured area is 7cm, then the attenuation factor = 5 ÷ 7 ≈ 0.71, and the correction point density value is 0.06 × 0.71 ≈ 0.043 points / cm². 2 This naturally reduces its weight.

[0140] Based on the embodiments provided in this application, the blank sub-regions within the circumscribed convex hull are detected and the effective area is calculated, avoiding the distortion of point density values ​​caused by uneven distribution of monitoring points, so that the point density can more realistically reflect the actual distribution density of monitoring points in the crack area; an inverse proportional attenuation factor is applied to the corrected point density value of the fractured area, which can reflect the integrity differences of the crack area in the subsequent weight allocation, allowing the risk assessment to distinguish the different degrees of harm between concentrated and dispersed crack distribution, and improving the fit between the diagnostic results and the actual crack morphology.

[0141] Furthermore, when spraying fiber-reinforced repair materials at a fiber density higher than the baseline density, penetration path optimization is performed, including:

[0142] A pseudo-random sequence is generated based on the difference between the target viscosity range of the repair material and the current viscosity, driving the fiber nozzle to vibrate in the normal plane; wherein, the vibration frequency is negatively correlated with the strain gradient modulus, and the amplitude is proportional to the fiber density reinforcement ratio;

[0143] The normal plane refers to a virtual plane perpendicular to the surface of the crack (e.g., the normal plane of a wall crack is a plane perpendicular to the wall). Based on the difference between the target viscosity and the actual viscosity (the larger the difference, the more obvious the deviation in material flowability), an irregular angular fluctuation sequence (pseudo-random sequence) is generated. The nozzle is controlled to vibrate within the normal plane according to this sequence to expand the spray coverage and adapt to changes in material flowability.

[0144] Real-time calculation of the divergence value of the spatial gradient vector of strain data within the area covered by the jet trajectory;

[0145] Among them, the strain gradient modulus reflects the stress concentration in the crack region (the larger the modulus, the more concentrated the stress), and the vibration frequency is negatively correlated with it, that is, the spraying frequency decreases at the stress concentration point (the spraying is slower), ensuring that the material is fully filled; the fiber density reinforcement ratio (a multiple of the base density) determines the amplitude, the higher the ratio, the larger the amplitude (the wider the coverage), to match the demand for high fiber usage.

[0146] The divergence value of the spatial gradient vector describes the convergence or divergence of strain within the spraying area (positive values ​​indicate convergence, negative values ​​indicate divergence). By calculating this value in real time, the permeability resistance of the repair material is determined—a larger divergence value (more pronounced convergence) indicates greater difficulty in material penetration into the area (potentially indicating the presence of impurities or air bubbles within the crack). For example, a significant convergence of the strain gradient vector within the sprayed area (divergence value of 0.06 / μm) indicates permeability resistance in the area, possibly due to loose concrete debris within the crack, making material filling difficult.

[0147] When the divergence value exceeds the penetration threshold, the vibration direction is rotated in the normal plane. The rotation angle is determined by mapping the derivative of the divergence value with respect to time through the hyperbolic tangent function. The rate of change of divergence after rotation is recorded. If the rate of change of divergence does not converge, the fiber density is triggered to increase in stages.

[0148] The penetration threshold is a critical value used to determine whether the penetration resistance is too high (e.g., 0.05 / μm for concrete cracks). When the divergence value exceeds the limit, the rate of change (derivative) of the divergence value over time is calculated and converted into a rotation angle of the vibration direction using a hyperbolic tangent function (the larger the rate of change, the larger the angle), allowing the nozzle to quickly avoid high-resistance areas. For example, with a penetration threshold of 0.05 / μm, when the divergence value reaches 0.07 / μm and the derivative is 0.03 / μm·s (resistance increases rapidly), a rotation angle of 15° is obtained through mapping using the hyperbolic tangent function. The nozzle rotates 15° in the normal plane, bypassing the high-resistance area.

[0149] Non-convergence of the divergence rate means that the divergence value continues to increase after rotation (resistance does not decrease). In this case, by progressively increasing the fiber density (e.g., increasing the base density by 50% in each stage), a "skeleton" is formed using high-fiber-content materials to overcome the penetration resistance (e.g., pushing away impurities within cracks), until the divergence rate stabilizes (converges). For example, if the divergence value increases from 0.07 / μm to 0.09 / μm after rotation (non-convergence), it triggers a progressive increase in fiber density from 150% → 200% → 250% of the base density until the divergence rate stabilizes within ±0.01 / μm·s. Based on the embodiments provided in this application, the fiber nozzle is driven to vibrate in the normal plane by a pseudo-random sequence, and the vibration frequency is negatively correlated with the strain gradient modulus and the amplitude is positively correlated with the fiber density enhancement ratio. This allows the fiber spraying coverage to better adapt to the spatial changes of the stress gradient and improve the uniformity of fiber distribution in complex crack areas. Adjusting the vibration direction and fiber density based on the strain gradient vector divergence value allows the spraying path to actively adapt to the differences in permeation resistance inside the crack, avoiding the problem of insufficient fiber accumulation in high-resistance areas and enhancing the tightness of the bond between the repair material and the crack.

[0150] Furthermore, while monitoring the recovery of resistance values ​​in the repaired area, a regional healing assessment is performed simultaneously, including:

[0151] The resistance recovery rate is calculated independently for each continuous region, and the rate value is the absolute value of the resistance change per unit time.

[0152] The monitoring data was divided into continuous areas, and the resistance recovery rate (the absolute change in resistance value per unit time) of each area was calculated independently. The purpose was to accurately locate areas with poor repair results, since the crack morphology and repair material distribution may differ in different areas, and independent calculation can prevent overall average data from masking local problems.

[0153] The region healing coefficient is defined as the product of the resistive recovery rate and the area of ​​the region's circumscribed convex hull.

[0154] The method comprehensively evaluates the repair effect of areas of different sizes by multiplying the resistance recovery rate (repair speed) by the area (repair scale). This avoids judging solely by rate (e.g., small areas have fast rates but small areas, while large areas have slow rates but large areas), making the repair quality of areas of different sizes comparable.

[0155] When the regional healing coefficient is lower than the median of the repaired regions in the same batch, a local reinforcement instruction is generated:

[0156] The median healing coefficient of the repaired areas in the same batch is used as a benchmark. Areas with a coefficient lower than this value are identified as weak repair areas (e.g., if the median is 120, a certain area has a coefficient of 80). The generated reinforcement command includes information such as the coordinates and centroid position of the area, ensuring that the reinforcement operation is accurately targeted at the weak area.

[0157] Add radial jetting paths at the centroid of the corresponding continuous region, with the number of paths equal to the integer part of the point density value of that region;

[0158] The process begins at the region's centroid (the geometric center of its circumscribed convex hull) and expands outwards to increase the number of injection paths. The number of paths is determined by the integer part of the point density value (higher density results in more paths). The goal is to distribute the reinforcing material along the direction in which the crack may extend, adapting to the spatial propagation characteristics of the crack.

[0159] The spraying time for each path is inversely proportional to the normalized value of the strain gradient modulus in that path direction;

[0160] The normalized value of the strain gradient modulus is the ratio of the modulus in that direction to the maximum modulus in the region (between 0 and 1). The spraying time is inversely proportional to this; that is, the larger the gradient modulus (the more concentrated the stress), the longer the spraying time, ensuring more complete material filling in the high-stress direction.

[0161] After reinforcement is completed, the regional healing coefficient balance is recalculated. If the regional healing coefficient balance exceeds the tolerance threshold, the preset connectivity threshold is optimized.

[0162] The regional healing coefficient uniformity refers to the dispersion (e.g., standard deviation) of the healing coefficients in each region, while the tolerance threshold is the maximum allowable dispersion range. If the uniformity exceeds the standard, it indicates that the regional division may be unreasonable (e.g., related regions are split). By optimizing the connectivity threshold (the distance standard for judging regional correlation), subsequent regional divisions can better reflect the actual correlation of cracks, thus improving the uniformity of repair. For example, if the standard deviation of the regional healing coefficient after reinforcement is 18 (tolerance threshold 12), optimizing the connectivity threshold from 7cm to 5cm and re-clustering reduces the standard deviation to 10 (meeting the tolerance), making the regional division more suitable for the actual distribution of cracks.

[0163] Based on the embodiments provided in this application, the resistance recovery rate and regional healing coefficient are calculated on a continuous region basis, which can accurately identify local areas with poor repair effects. Radial spray paths are added at the centroid, and the number of paths and spray duration are related to point density and strain gradient characteristics, so that the reinforcement operation can specifically compensate for the weak parts and improve the uniformity of regional repair effects. By optimizing the preset connectivity threshold to improve the uniformity of regional healing coefficients, the subsequent regional division can be more adapted to the actual needs of repair operations, and the adaptability of the method in repeated repair scenarios can be enhanced.

[0164] It should be noted that the embodiments implemented on the self-diagnosis and self-repair system side of building structure cracks in this application can be referenced with the embodiments implemented on the self-diagnosis and self-repair method side of building structure cracks, and will not be described in detail here.

[0165] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described xx method is also provided, the electronic device being... Figure 4 The terminal device or server shown. This embodiment uses this electronic device as an example of a server. Figure 4 As shown, the electronic device includes a memory 402, a processor 404, and a transmission device 406. The memory 402 stores a computer program, and the processor 404 is configured to execute the steps in any of the above method embodiments through the computer program.

[0166] Optionally, in this embodiment, the aforementioned electronic device may be located in at least one of a plurality of network devices in a computer network.

[0167] Optionally, the transmission device 406 is used to receive or send data via a network. Specific examples of the network described above may include wired and wireless networks. In one example, the transmission device 406 includes a Network Interface Controller (NIC), which can be connected to other network devices and a router via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 406 is a radio frequency (RF) module used to communicate with the Internet wirelessly.

[0168] In addition, the aforementioned electronic device also includes: a display 408 for displaying target identification characters contained in the identity identifier of the identified target object; and a connection bus 410 for connecting various module components in the aforementioned electronic device.

[0169] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A self-diagnosis and self-repair system for building structural cracks, characterized in that, include: The crack data acquisition module is used to synchronously acquire the resistance time series data, strain spatial distribution data and environmental erosion parameters of the cracks through a pre-embedded sensor array when cracks appear in the building structure. The target monitoring point identification module is used to calculate the change in the resistance time series data per unit time as the expansion rate value, and simultaneously scan the strain spatial distribution data to identify the coordinates of the target monitoring points where the strain value change rate exceeds the critical point; wherein, the strain value change rate is the absolute value of the strain difference between adjacent monitoring points divided by the distance between monitoring points. The point density value determination module is used to cluster target monitoring points whose Euclidean distance between them is less than a preset aggregation threshold into continuous regions based on the coordinates of the identified target monitoring points; and to calculate the ratio of the number of target monitoring points included in each continuous region to the area of ​​the circumscribed convex hull of the region as the point density value of the region. The risk determination module is used to dynamically calculate weights for the propagation rate value, the point density value, and the environmental erosion parameter based on the cumulative time after the crack appears, and then sum the three weighted values ​​to determine the comprehensive risk value. The repair module is used to control the repair equipment to perform repair operations based on the comprehensive risk value.

2. A self-diagnosis and self-repair method for cracks in building structures, characterized in that, include: When cracks appear in the building structure, the resistance time series data, strain spatial distribution data and environmental erosion parameters of the cracks are collected synchronously through a pre-embedded sensor array. The change in the resistance time series data per unit time is calculated as the expansion rate value. At the same time, the strain spatial distribution data is scanned to identify the coordinates of the target monitoring points where the strain value change rate exceeds the critical point. The strain value change rate is the absolute value of the strain difference between adjacent monitoring points divided by the distance between the monitoring points. Based on the coordinates of the identified target monitoring points, target monitoring points whose Euclidean distance to each other is less than a preset aggregation threshold are clustered into continuous regions; the ratio of the number of target monitoring points in each continuous region to the area of ​​the circumscribed convex hull of the region is calculated as the point density value of the region. Based on the cumulative duration after the crack appears, the propagation rate value, the point density value, and the environmental erosion parameter are dynamically weighted and then weighted and summed to determine the comprehensive risk value. The repair equipment is controlled to perform repair operations based on the comprehensive risk value.

3. The self-diagnosis and self-repair method for building structural cracks according to claim 2, characterized in that, The step of controlling the repair equipment to perform repair operations based on the comprehensive risk value includes: The crack risk level is determined by matching the comprehensive risk value with a preset risk level table, and the point density values ​​are sorted from largest to smallest. The continuous area with the first preset proportion is marked as a high stress concentration area. Based on the crack risk level, the target viscosity range of the repair material is determined by querying a preset viscosity mapping table, and the coordinate range of the high stress concentration area is mapped to the fiber density reinforced region. Controlling the repair equipment to perform repair operations includes: preparing a repair material matrix according to the target viscosity range of the repair material; spraying fiber-reinforced repair material at a fiber density higher than the base density in the fiber density-enhanced region; and calculating the spatial gradient vector of strain data in the high stress concentration area in real time, and controlling the fiber nozzle to spray in the opposite direction of the strain gradient.

4. The self-diagnosis and self-repair method for building structural cracks according to claim 3, characterized in that, After the repair equipment is controlled to perform repair operations based on the comprehensive risk value, the method further includes: Monitor the recovery of resistance in the repaired area. If the amount of resistance recovery per unit time is lower than the recovery threshold set according to the crack risk level during the diagnostic stage, increase the flow rate of the curing agent. After the repair material has cured, a standard test load is applied to the cured area. The maximum fluctuation of the resistance value under the load and the reciprocal of the time required for the strain value to recover to the initial value as a set percentage after unloading are collected as the strain recovery rate. Based on the maximum fluctuation range of the resistance value and the strain recovery rate, the weight distribution ratio of the expansion rate value, the point density value, and the environmental erosion parameter is readjusted, and the preset risk level table and the preset viscosity mapping table are updated.

5. The self-diagnosis and self-repair method for building structural cracks according to claim 4, characterized in that, The step of assigning dynamic calculation weights to the propagation rate value, the point density value, and the environmental erosion parameter based on the cumulative duration after the crack appears includes: A time-efficiency function for the crack development stage is constructed, and the cumulative duration is mapped to a stage coefficient, wherein the weight of the expansion rate value decreases according to a hyperbolic tangent curve as the stage coefficient increases. For the point density value, the spatial topological connectivity between each continuous region is calculated, including: if the Euclidean distance between the boundary points of two regions is less than a preset connectivity threshold, then a connection edge is established, and the ratio of the number of connection edges to the total number of regions is the regional network density. Multiply the point density value by the natural logarithm of the regional network density to generate a topology-weighted density index as the weight input; The absolute value of the environmental erosion parameters is integrated using a sliding time window, and then convolved with the stage coefficients to generate the cumulative erosion intensity. Dynamic weights are generated based on a combination of the hyperbolic tangent attenuation weight controlled by the stage coefficient, the topological weighted density index, and the cumulative erosion intensity.

6. The self-diagnosis and self-repair method for building structural cracks according to claim 2, characterized in that, The control of the fiber nozzle to spray in the reverse strain gradient direction includes: A Delaunay triangular control mesh is constructed using the geometric centroids of each continuous region within the high stress concentration zone as control nodes. The strain gradient vector at the vertex of each triangular element is calculated in real time, and the maximum value of the vector magnitude within the element is taken as the reference direction. The difference between the median of the target viscosity range of the repair material and the current viscosity is mapped to a directional perturbation angle using a hyperbolic tangent function. The theoretical injection direction is generated by superimposing the disturbance angle in the opposite direction of the reference direction, while the second spatial derivative of the strain energy density in the theoretical direction projection region is scanned. If the second spatial derivative exceeds the curvature threshold, the injection angle is deflected along the normal of the strain energy density contour line, and the amount of deflection is proportional to the rate of curvature change.

7. The self-diagnosis and self-repair method for building structural cracks according to claim 5, characterized in that, The weighting ratio is readjusted based on the maximum fluctuation range of the resistance value and the strain recovery rate, including: The maximum fluctuation range of the resistance value is converted into a structural stiffness reduction factor, and the reciprocal of the strain recovery rate is converted into a material relaxation time constant. A five-dimensional state space is established, comprising the expansion rate value, the point density value, the environmental erosion parameter, the structural stiffness reduction coefficient, and the material relaxation time constant. The dominant feature plane is extracted through recursive orthogonal decomposition. Constructing a weight correction vector within the dominant feature plane includes: a spread rate weight correction term proportional to the projection of the structural stiffness reduction coefficient onto the basis vector of the dominant feature plane; The point density weight correction term implements a region fusion mechanism, including: when the difference in relaxation time constants between two consecutive regions is less than the fusion tolerance, merging the regions and recalculating the circumscribed convex hull area and topology weighted density index; The environmental erosion weight correction term introduces an erosion-stiffness coupling factor, which is the absolute value of the partial derivative of the structural stiffness reduction coefficient with respect to the cumulative erosion intensity. The comprehensive risk value is recalculated using the updated continuous regions and corrected weights, and the preset aggregation threshold and preset connectivity threshold are optimized simultaneously to minimize the variance of the number of merged regions.

8. The self-diagnosis and self-repair method for building structural cracks according to claim 2, characterized in that, When calculating the ratio of the number of target monitoring points within each continuous region to the area of ​​the circumscribed convex hull of that region, a region integrity correction is performed, including: Detect blank sub-regions without target monitoring points inside the circumscribed convex hull, and calculate the proportion of the area of ​​the blank sub-region to the total area of ​​the convex hull as the blank ratio; Extract the target monitoring point that is closest to the boundary of the blank sub-region and calculate its physical Euclidean distance to the geometric center of the blank sub-region; When the blank area ratio exceeds the area tolerance and the physical Euclidean distance exceeds the displacement tolerance, the area of ​​the blank sub-region is subtracted from the total area of ​​the convex hull as the effective calculated area. The corrected point density value is obtained by dividing the number of target monitoring points by the effective calculated area, and the continuous area that triggers the correction is recorded and marked as the broken area. When assigning dynamic weights to the point density values ​​in the subsequent process, an inverse proportional attenuation factor based on displacement tolerance is applied to the corrected point density values ​​of the broken regions.

9. The self-diagnosis and self-repair method for building structural cracks according to claim 3, characterized in that, When spraying the fiber-reinforced repair material at a fiber density higher than the baseline density, penetration path optimization is performed, including: A pseudo-random sequence is generated based on the difference between the target viscosity range of the repair material and the current viscosity, driving the fiber nozzle to vibrate in the normal plane; wherein, the vibration frequency is negatively correlated with the strain gradient modulus, and the amplitude is proportional to the fiber density reinforcement ratio; Real-time calculation of the divergence value of the spatial gradient vector of strain data within the area covered by the jet trajectory; When the divergence value exceeds the penetration threshold, the vibration direction is rotated in the normal plane. The rotation angle is determined by mapping the derivative of the divergence value with respect to time through the hyperbolic tangent function. The rate of change of divergence after rotation is recorded. If the rate of change of divergence does not converge, the fiber density is triggered to increase in stages.

10. The self-diagnosis and self-repair method for building structural cracks according to claim 5, characterized in that, While monitoring the recovery of resistance values ​​in the repaired area, a regional healing assessment is performed simultaneously, including: The resistance recovery rate is calculated independently for each continuous region, and the rate value is the absolute value of the resistance change per unit time. The region healing coefficient is defined as the product of the resistive recovery rate and the area of ​​the region's circumscribed convex hull. When the healing coefficient of the area is lower than the median of the repaired areas in the same batch, a local reinforcement instruction is generated: Add radial jetting paths at the centroid of the corresponding continuous region, with the number of paths equal to the integer part of the point density value of that region; The spraying time for each path is inversely proportional to the normalized value of the strain gradient modulus in that path direction; After reinforcement is completed, the regional healing coefficient balance is recalculated. If the regional healing coefficient balance exceeds the tolerance threshold, the preset connectivity threshold is optimized.

Images