Methods for controlling seepage in high-density shotcrete for steel structure floor slabs on irregularly shaped roofs

Abstract

This invention discloses a method for controlling the impermeability of high-density shotcrete for steel structure floor decks on irregularly shaped roofs, belonging to the field of concrete impermeability control technology. The method includes the following steps: During the concrete spraying process, the rebound trajectory field and shear disturbance trajectory field are simultaneously collected at the marked convex curved surface transition zone. Combined with the jet velocity disturbance spectrum data, the change in the reverse scouring pressure distribution in the convex curved surface transition zone during concrete spraying is determined, forming a disturbance energy density distribution map. The disturbance energy density distribution map is then linked with the surface response curve of the concrete at the initial setting stage for calculation, establishing a density weakening prediction factor model for assessing the integrity of the sprayed layer, and outputting a weakening risk index. This invention solves the problem of difficulty in identifying density weakening of the sprayed layer in the convex curved surface transition zone due to secondary scouring, achieving real-time monitoring and closed-loop control during the spraying process, and ensuring impermeability quality.

Description

Technical Field

[0001] This invention relates to the field of concrete impermeability control technology, specifically to a method for impermeability control of high-density shotcrete for steel structure floor decking of irregular roofs. Background Technology

[0002] High-density shotcrete for waterproofing steel structure floor decking in irregularly shaped roofs mainly refers to the use of specially designed shotcrete construction techniques and processes in complex geometric roof structures to create a high-density bonding interface between the concrete and the steel structure floor decking, thereby improving the overall waterproofing capacity of the structure. Existing technologies typically employ high-pressure spraying equipment to apply a specific mix of high-performance concrete along the surface of the steel structure floor decking at multiple angles and in multiple layers, supplemented by interface treatment agents or bonding reinforcement materials, to address issues such as uneven concrete adhesion, insufficient density, and leakage caused by irregular curved surfaces. This process typically includes the following key steps: First, the surface of the irregularly shaped steel structure floor deck is cleaned and pre-treated, such as by grinding and removing rust and applying an interface agent, to enhance the initial adhesion of the concrete. Second, a first base coat is sprayed using a specific spraying angle and path to ensure complete coverage of the structural outline. Then, the main layer is sprayed in layers as needed, with density and uniformity tests performed between each layer. Finally, finishing, compaction, and curing further enhance the overall density and impermeability of the shotcrete. Throughout the process, controlling the spraying pressure, concrete composition, spraying thickness, and construction intervals are crucial technical means to achieve a high-density, impermeable effect, thereby ensuring the structural stability and waterproof performance of the irregularly shaped roof in harsh environments.

[0003] The existing technology has the following shortcomings:

[0004] When using high-density shotcrete for waterproofing control of irregularly shaped steel structure floor decks, if the spraying area is located in the convex curved transition zone formed by the steel structure connection, the spraying pressure will be concentrated and superimposed on the transition surface due to the abrupt curvature change at this location. In such cases, the shotcrete particles will generate rebound waves and shear disturbances after impacting the surface at high speed, which can easily cause "secondary scouring" of the already initially set sprayed layer. This weakens the originally established dense structure and produces loosening or micropore defects that are difficult to visually identify. Existing high-density shotcrete waterproofing control technology for irregularly shaped steel structure floor decks cannot determine whether the density of the already formed sprayed layer in the convex curved transition zone has been damaged based on the changes in the reverse scouring pressure distribution during the concrete spraying process. This makes it impossible to adjust the spraying method or parameters in time during construction, resulting in potential weak points in the local sprayed layer. Consequently, under the action of subsequent water flow or wind pressure, this can lead to structural continuity failure and large-scale waterproofing failure, which is a hidden waterproofing quality problem.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method for controlling the seepage of high-density shotcrete for steel structure floor decking on irregularly shaped roofs, so as to solve the problems in the background art mentioned above.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for controlling the impermeability of high-density shotcrete for steel structure floor decking of irregularly shaped roofs, specifically including the following steps:

[0008] S1. Perform three-dimensional structural analysis on the steel structure floor deck of the irregular roof. By extracting the positive gradient abrupt increase boundary in the continuous curvature change in space, identify the convex surface transition zone and mark the curvature change zone in the convex surface transition zone as the subsequent monitoring area.

[0009] S2. During the concrete spraying process, the rebound trajectory field and shear disturbance trajectory field are simultaneously collected in the marked convex curved surface transition area. Combined with the spray jet velocity disturbance spectrum data, the change in the reverse scouring pressure distribution in the convex curved surface transition area during the concrete spraying process is determined, and a disturbance energy density distribution map is formed.

[0010] S3. Link the disturbance energy density distribution map with the surface response curve of the initial setting stage of concrete to establish a density weakening prediction factor model for evaluating the integrity status of the sprayed layer, and output the weakening risk index.

[0011] S4. By combining the risk reduction index with the jet trajectory time series data and the jet beam effect distribution results, nonlinear fitting is performed to determine whether the density of the formed spray layer in the convex curved surface transition zone is damaged, and the spray layer structure state is divided into multiple levels.

[0012] S5. Based on the classification of the spray layer structure, the spray energy density, spray angle, path speed and mixture ratio are adjusted in a coordinated manner, and the disturbance energy density distribution map is updated in real time to correct the density reduction prediction factor model, so as to realize the dynamic closed-loop control of the spray layer density.

[0013] Preferably, S1 is as follows:

[0014] A three-dimensional structural model containing node spatial coordinates and node normal vectors is constructed for the irregular roof steel structure floor deck. By calculating the curvature sequence of the node normal vectors, the three-dimensional structural unfolding and analysis of the irregular roof steel structure floor deck is completed, and the spatial continuous curvature change distribution is obtained.

[0015] In the distribution of continuous curvature in space, the positive gradient of curvature with spatial position is calculated to extract the positive gradient abrupt boundary. The curvature abrupt boundary set is established at the position where the positive gradient shows continuous abrupt increase, which is used to identify candidate convex surface regions that show abrupt increase characteristics in the continuous curvature distribution in space.

[0016] Based on the rotational aggregation trend of the node normal vector and the continuity of the curvature growth direction, the convex surface transition zone is identified in the curvature abrupt increase boundary concentration. Then, the curvature abrupt increase boundary of the convex surface transition zone is used as the center to expand outward to the curvature change stable region to calibrate the curvature change abrupt zone, thus completing the establishment of the monitoring area for the convex surface transition zone.

[0017] Preferably, S2 specifically includes the following steps:

[0018] S201. During the concrete spraying process, a synchronous imaging acquisition array is set up in the calibrated convex curved surface transition area to continuously record the rebound path formed after the concrete particles are impacted and the surface tangential displacement caused by the spraying disturbance. The rebound trajectory field and shear disturbance trajectory field are obtained by reconstructing through a unified time axis.

[0019] S202. Synchronously align the disturbance amplitude and jet velocity disturbance spectrum data of the corresponding points in the rebound trajectory field and shear disturbance trajectory field point by point, extract the rebound direction change rate, shear displacement response amplitude and spectrum disturbance energy characteristic value at each spatial position, and construct the jet disturbance parameter fusion matrix.

[0020] S203. Based on the weighted coupling relationship between the disturbance eigenvalues ​​at each position in the fusion matrix of spraying disturbance parameters, calculate the unit disturbance energy of each spatial point in the convex surface transition zone, and generate a disturbance energy density distribution map with continuous spatial coordinates as index, which is used to determine the change in reverse scouring pressure distribution in the convex surface transition zone during concrete spraying.

[0021] Preferably, S203 is as follows:

[0022] The rebound direction change rate, shear displacement response amplitude, and jet velocity disturbance spectrum intensity value of each spatial point are extracted from the jet disturbance parameter fusion matrix. A combined correlation table between each group of ternary disturbance feature values ​​is constructed. The local correlation coefficient is calculated by the convolution kernel sliding window method, and a position disturbance feature weight mapping map is generated accordingly.

[0023] Based on the weight coupling relationship of perturbation features at different locations in the location perturbation feature weight mapping map, a weighted integral operation is performed to calculate the unit perturbation energy corresponding to each spatial point. The unit perturbation energy distribution body is formed according to the three-dimensional coordinate order of the spatial points, and a spatial perturbation energy index model is constructed.

[0024] Based on the spatial disturbance energy index model, the gradient flow direction and density accumulation degree of disturbance energy in space are calculated. Based on the abrupt change in the slope of disturbance energy change, the abnormal concentration area of ​​reverse scour pressure is identified, and a disturbance energy density distribution map is generated to determine the change in reverse scour pressure distribution in the convex curved surface transition area during concrete spraying.

[0025] Preferably, S3 specifically includes the following steps:

[0026] S301. Align the disturbance energy value of each spatial point in the disturbance energy density distribution map with the surface response curve collected at the corresponding spatial point during the initial setting stage of concrete, and extract the rate of change of disturbance energy change, response delay interval and response recovery time period in the response curve as linkage input variables.

[0027] S302. Based on the linked input variables, construct a disturbance response coupled dataset, use linear discriminant clustering and gradient regression screening algorithms to extract key parameter sets that reflect the characteristics of spray layer density weakening, establish a density weakening prediction factor model for evaluating the integrity status of the spray layer, and generate density weakening prediction factor values ​​for each spatial point.

[0028] S303. The density weakening prediction factor value is mapped in an orderly manner according to the spatial point coordinates, and the local gradient change trend of the disturbance energy density distribution map is superimposed to calculate the overall weakening degree of the spray layer and generate a weakening risk index, which is used to reflect the spatial distribution of the integrity state of the spray layer.

[0029] Preferably, S302 is as follows:

[0030] A perturbation response coupled dataset is constructed using the perturbation energy density distribution map and the perturbation energy value, response amplitude change rate, response delay interval and response recovery time period at the corresponding spatial location in the surface response curve as input, and the input variables are bound to the spray layer location using spatial coordinates as index;

[0031] Linear discriminant clustering was performed on the perturbation-response coupling dataset to extract data clusters that exhibited the coupling behavior of abrupt changes in perturbation peaks and abnormal responses. Then, a gradient regression screening algorithm was used to remove weak interfering variables and select key parameter groups that reflect the weakening state of the spray layer's compactness.

[0032] The key parameter set is input into the multi-factor comprehensive evaluation framework. The peak sensitivity of the disturbance, the hysteresis degree of the response curve and the spatial consistency are used as evaluation indicators to construct a compaction weakening prediction factor model. The compaction weakening prediction factor value is calculated for each spatial point for spray layer integrity status assessment.

[0033] Preferably, S303 is as follows:

[0034] The compactness reduction prediction factor values ​​are arranged in an orderly manner according to the three-dimensional coordinates of spatial points, and a compactness reduction state field is constructed under a unified coordinate system to realize the continuous mapping of the reduction factor in space.

[0035] Based on the compactness weakening state field, the local gradient change trend of the corresponding spatial point in the perturbation energy density distribution map is used as an incremental weight factor, and it is superimposed and fused with the compactness weakening prediction factor value point by point to construct a multi-dimensional perturbation-compactness comprehensive evaluation matrix.

[0036] Local integration calculations are performed on the multidimensional perturbation-density comprehensive evaluation matrix. The overall degree of spray layer weakening is calculated by combining the spatial point clustering trend. Based on the set grading standard, a weakening risk index is output to characterize the spatial distribution characteristics of the spray layer integrity state.

[0037] Preferably, S4 is as follows:

[0038] The index value of each spatial point in the weakening risk index is aligned one-to-one with the trajectory displacement curve of the corresponding spatial point in the jet trajectory time series data, as well as the jet intensity and jet duration of the corresponding spatial point in the jet effect distribution results according to the three-dimensional spatial coordinates. A nonlinear fitting input dataset indexed by three-dimensional spatial coordinates is constructed, and amplitude standardization is performed on all input variables to ensure the stability of the fitting operation.

[0039] Based on the nonlinear fitting input dataset, a multidimensional nonlinear fitting algorithm is used to establish a response fitting model between the risk reduction index and the jet trajectory time series data and the jet effect distribution results, and the fitting residual value and response offset of each spatial point are calculated. When the fitting residual value of a spatial point exceeds the fitting residual threshold and its response offset shows a continuous increasing trend, it is determined that the compactness of the formed spray layer in the convex curved surface transition area corresponding to the spatial point is damaged, and the spatial point is added to the set of compactness damage points.

[0040] Based on the fitting residual amplitude, response offset, and degree of aggregation of spatial points in the set of dense failure points, a density grading standard is set, and the structural state of the formed sprayed layer in the convex curved surface transition zone is divided into multiple levels: dense and stable, slightly weakened, moderately weakened, and severely weakened, and a spatial level distribution map of the sprayed layer structural state is formed.

[0041] Preferably, S5 is as follows:

[0042] Based on the division of the spray layer structure state, a parameter control mapping relationship is established. Different density level regions are matched with spray energy density, spray angle, path speed and mixture ratio parameters to generate a spray control target set and bind it to the structural state space coordinates to indicate the process adjustment strategies required for different spatial points.

[0043] Based on the injection energy density output module, nozzle angle control module, path motion control module and mixing ratio supply module of the injection control target set linkage control injection system, the system collects the disturbance response data generated during the injection process in real time while performing differentiated injection operations, and calculates the disturbance energy parameters in the new round of injection intervention area online.

[0044] The updated disturbance energy parameters in the injection intervention area are fused with the original disturbance energy density distribution map to construct a dynamic update map of disturbance energy density. The dynamic update map is then used as input to adjust the weights of characteristic variables in the density reduction prediction factor model in real time, forming a linkage mapping model between disturbance control and density assessment, thereby achieving dynamic closed-loop control of the spray layer density.

[0045] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0046] 1. This invention addresses the problem of spray layer density damage caused by abrupt curvature changes in the transition zone of a convex surface, constructing a closed-loop system encompassing monitoring, identification, evaluation, and control throughout the entire process. This technical solution accurately identifies curvature change zones through three-dimensional structural analysis, achieving multi-dimensional data fusion of disturbance energy and surface response during spraying. This leads to the construction of a disturbance response coupled dataset and a density reduction prediction factor model, which not only quantitatively assesses the integrity of the spray layer but also outputs a spatially distributed reduction risk index. This enables proactive early warning and precise location of potential density defects, avoiding the judgment errors caused by the lag and subjectivity of traditional visual assessment.

[0047] 2. This invention constructs a graded model of the sprayed layer structure state through nonlinear fitting and a density failure point identification mechanism. The identification results are then linked inversely to the spraying parameter control module to dynamically adjust the spraying energy density, angle, path, and material ratio. By leveraging real-time updates of the disturbance energy density and continuous correction of the predictive factor model, the system achieves tightly coupled closed-loop control between disturbance monitoring and density assessment. This highly integrated data-driven and process-linked mechanism improves the compactness, uniformity, and impermeability stability of the sprayed layer in convex curved areas, preventing waterproofing failures caused by localized weakening and ensuring the quality continuity and structural durability of concrete spraying construction in complex spatial structures. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0049] Figure 1This is a schematic flowchart of the method for controlling the seepage of high-density shotcrete used in steel structure floor decking for irregular roofs according to the present invention. Detailed Implementation

[0050] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0051] This invention provides, for example Figure 1 The method for controlling the seepage of high-density shotcrete for steel structure floor decking on irregularly shaped roofs, as shown, specifically includes the following steps:

[0052] S1. Perform three-dimensional structural analysis on the steel structure floor deck of the irregular roof. By extracting the positive gradient abrupt increase boundary in the continuous curvature change in space, identify the convex surface transition zone and mark the curvature change zone in the convex surface transition zone as the subsequent monitoring area.

[0053] In this embodiment, S1 specifically refers to:

[0054] A three-dimensional structural model containing node spatial coordinates and node normal vectors is constructed for the irregular roof steel structure floor deck. By calculating the curvature sequence of the node normal vectors, the three-dimensional structural unfolding and analysis of the irregular roof steel structure floor deck is completed, and the spatial continuous curvature change distribution is obtained.

[0055] To construct a 3D structural model of the irregularly shaped roof steel structure floor deck, including the spatial coordinates and normal vectors of the nodes, the spatial geometric information of the floor deck needs to be acquired first. This can be done by creating a high-precision digital surface model using a 3D laser scanner or structural modeling software (such as Revit or Rhino). The model represents the structural shape as a curved mesh, where each mesh node contains its position coordinates in 3D space and the normal direction of the surface it lies on. To obtain the curvature variation of the structural surface in different regions, the normal vector of each node needs to be continuously compared, and the trend of the change in the normal angle between adjacent nodes needs to be calculated, thus constructing a sequence of curvature variation with spatial distribution. This process can reveal which regions of the structural surface experience abrupt geometric transitions, and is particularly suitable for detecting sudden convexity transitions in curved surfaces. By establishing a complete and continuous spatial curvature variation distribution in this way, it is helpful for subsequent prediction of the concentration trend of spray pressure, thus providing a theoretical basis for monitoring the density of the sprayed layer.

[0056] Three-dimensional structural models are geometric representations of surfaces generated from discrete spatial points, typically constructed using polygonal meshes (such as triangular meshes), where nodes are the basic units representing the structural outline. Each node's spatial coordinates refer to its position in three-dimensional space, usually stored in X, Y, and Z format; the node normal vector is the direction perpendicular to the surface of the face containing the node, used to represent the local orientation of the surface at that point. Curvature sequence calculations are derived by analyzing the changes in normal vectors between consecutive nodes. For example, in a certain curved surface region, if the change in the node normal vector shows a rapid deflection over a short distance, it indicates a geometric abrupt change, with relatively higher curvature. For instance, in a gently curving arc region, the normal vector changes slowly, while at a convex corner, the change is very drastic. Curvature sequences are used to quantify this trend, providing a computational basis for identifying high-risk transition zones. Through complete modeling and continuous curvature analysis, a geometrical perception of the entire irregular structure surface can be formed, effectively guiding subsequent anti-seepage monitoring strategies.

[0057] In the distribution of continuous curvature in space, the positive gradient of curvature with spatial position is calculated to extract the positive gradient abrupt boundary. The curvature abrupt boundary set is established at the position where the positive gradient shows continuous abrupt increase, which is used to identify candidate convex surface regions that show abrupt increase characteristics in the continuous curvature distribution in space.

[0058] In a spatially continuous curvature distribution, calculating the positive growth gradient of curvature with spatial location can effectively capture the trend of a structural surface transitioning from a gentle region to a sharply curved region. This process involves extracting a curvature sequence along a specific direction of the spatial surface and then calculating the increase in curvature values ​​between adjacent nodes, i.e., the positive growth gradient. By traversing the entire curvature distribution of the structural surface, locations where the curvature growth trend significantly intensifies in a certain direction are identified, and the sustainability of this growth is determined. For example, if three or more consecutive nodes show a gradual increase in the rate of curvature growth, this change represents a continuous abrupt increase in the positive growth gradient. These locations typically indicate areas prone to pressure superposition or rebound disturbances during spraying operations. Combining these locations of continuous abrupt increases forms a curvature abrupt increase boundary set, used to delineate the spatial boundary range where curvature abrupt changes are significant. This process helps to identify sensitive areas where the sprayed layer may experience dense breakdown in advance, thus providing precise target area location information for subsequent dense monitoring and control.

[0059] A positive gradient abrupt increase boundary refers to a boundary line along a spatial surface where the curvature value changes rapidly at consecutive nodes, and the rate of increase itself intensifies. This boundary is not based on the absolute magnitude of the curvature value, but rather on the abrupt change in the rate of increase, making it more suitable for identifying structural features transitioning from smooth to highly convex surfaces. The location of a continuous abrupt increase in the positive gradient refers to a situation where curvature growth not only exists in the curvature distribution sequence, but the magnitude of the increase also continuously increases at multiple consecutive nodes, exhibiting a second-order progression of the growth gradient. This phenomenon is an early signal of the formation of a convex surface transition. For example, at the connection point transitioning from a horizontal floor to an arched truss, curvature growth is not instantaneous but rather a gradual process of increasing in size and steepening, easily causing concentrated jetting and scouring disturbances. An abrupt increase is a defining node where a relative quantitative change transforms into a qualitative change, and is a key indicator for judging whether continuous curvature changes have broken through the stable zone. By establishing a set of abrupt increase boundaries, truly engineering-risk spatial areas in continuous changes can be identified from the overall structure, enabling targeted anti-seepage assessment and intervention deployment.

[0060] Based on the rotational aggregation trend of the node normal vector and the continuity of the curvature growth direction, the convex surface transition zone is identified in the curvature abrupt increase boundary concentration. Then, the curvature abrupt increase boundary of the convex surface transition zone is used as the center to expand outward to the curvature change stable region to calibrate the curvature change abrupt zone, thus completing the establishment of the monitoring area for the convex surface transition zone.

[0061] In the context of curvature abrupt increase boundaries, the key to identifying convex surface transition zones lies in determining the rotational convergence trend of nodal normal vectors in space and the continuity of the curvature growth direction. In practice, the normal vector of each node within the curvature abrupt increase boundary is first extracted, and a local normal rotation field is constructed by comparing its angle with the normals of neighboring nodes. When these normal vectors gradually converge and appear to converge around a certain spatial bend point, it can be determined that there is a normal rotational convergence trend in this region. Combined with the continuous extension of the curvature growth direction in this region, it is further confirmed whether the convergence center is located within a continuously increasing curvature channel. If both match, it indicates that this location is a high-risk point where the jet stream is prone to geometric disturbance accumulation, i.e., a convex surface transition zone. Subsequently, taking the center of this convex surface transition zone as the origin, the spatial region is expanded outwards until the curvature change gradually stabilizes, i.e., the curvature growth rate at continuous nodes decreases to within the fluctuation threshold, thus completing the calibration of the curvature abrupt change zone. Establishing monitoring areas in this way can maximize coverage of potentially sensitive areas where density degradation can occur due to abrupt changes in geometric shape.

[0062] The rotational convergence trend of nodal normal vectors refers to the consistency or convergence of the normal directions of multiple adjacent nodes on a structural surface in space. This means that these normal vectors converge or rotate around the same turning trend in a certain spatial region. This trend implies a stable geometric turn of the surface within that region, corresponding to convex deformation in physical space. The continuity of curvature growth direction refers to the continuous increase in curvature along a specific direction within the normal rotational convergence region, without any abrupt breaks or reversals. The superposition of these two characteristics indicates that the geometric changes in this region are not only directional but also continuous, representing the geometric root of abrupt changes in stable surfaces. A stable curvature change region refers to a region where the rate of change of curvature values ​​between nodes with position tends to be constant or the change is minimal, usually indicating that the structural morphology has transitioned from an abrupt change to a stable state. For example, in a curved roof ridge region, a large increase in curvature occurs around the central turning point, but as it continues to extend along the roof, the curvature tends to moderate, indicating that it has escaped the highly sensitive disturbance zone. By expanding outwards into regions with stable curvature changes, it is helpful to construct a more buffer-like anti-seepage monitoring coverage area, thereby improving identification accuracy and response stability.

[0063] S2. During the concrete spraying process, the rebound trajectory field and shear disturbance trajectory field are simultaneously collected in the marked convex curved surface transition area. Combined with the spray jet velocity disturbance spectrum data, the change in the reverse scouring pressure distribution in the convex curved surface transition area during the concrete spraying process is determined, and a disturbance energy density distribution map is formed.

[0064] In this embodiment, S2 specifically includes the following steps:

[0065] S201. During the concrete spraying process, a synchronous imaging acquisition array is set up in the calibrated convex curved surface transition area to continuously record the rebound path formed after the concrete particles are impacted and the surface tangential displacement caused by the spraying disturbance. The rebound trajectory field and shear disturbance trajectory field are obtained by reconstructing through a unified time axis.

[0066] In concrete spraying, to accurately capture the local disturbance response generated when sprayed particles interact with the structural surface, a synchronous imaging array composed of multiple high-speed vision units can be deployed around the pre-calibrated convex surface transition zone. Each vision unit is equipped with a fixed focal length lens and samples at a uniform frame rate. A clock synchronization mechanism ensures that the data frames captured by all units are aligned on the same time reference. Using particle impact points and surface disturbance traces in continuous image frames, the rebound path of sprayed particles and the tangential disturbance trajectory of the surface can be constructed respectively. To achieve temporal fusion of data, a unified time axis reconstruction method is required. This involves remapping the heterogeneous data collected by multiple vision units to a global time coordinate system using a time-stamped synchronization algorithm, thereby forming a complete and temporally continuous rebound trajectory field and shear disturbance trajectory field. This process enables dynamic reconstruction of the spraying disturbance response behavior of complex geometric surfaces and provides continuous, spatially well-distributed disturbance information for subsequent reverse scour identification.

[0067] Synchronous imaging acquisition arrays refer to array structures composed of multiple time-synchronized high-speed visual sensing units. Their design aims to simultaneously monitor the jetting reaction process in the transition zone of convex curved surfaces from multiple perspectives, avoiding data loss due to single-view occlusion or illumination deviations. Unified timeline reconstruction refers to correcting and interpolating the data frames captured by each visual unit to ensure that disturbance data from multiple spatial locations are synchronized at the same time node, guaranteeing the consistency and continuity of dynamic analysis results. The rebound trajectory field is a three-dimensional representation of the rebound path of jetting particles after impact, reflecting the direct correspondence between particle impact intensity and surface reaction. The shear disturbance trajectory field, on the other hand, maps the tangential displacement changes on the surface caused by jetting impact, appearing as a distribution image of minute slippage or micro-vibrations between surface particle layers. The combination of these two trajectory fields forms the basic sensing model of jetting disturbance, a crucial prerequisite data foundation for identifying changes in the distribution of reverse scour pressure.

[0068] S202. Synchronously align the disturbance amplitude and jet velocity disturbance spectrum data of the corresponding points in the rebound trajectory field and shear disturbance trajectory field point by point, extract the rebound direction change rate, shear displacement response amplitude and spectrum disturbance energy characteristic value at each spatial position, and construct the jet disturbance parameter fusion matrix.

[0069] To achieve in-depth analysis of the jet disturbance process, it is necessary to synchronize the disturbance amplitude and jet velocity disturbance spectrum data at the same spatial location in the rebound trajectory field and shear disturbance trajectory field point by point. Specifically, firstly, a unique index is established for each three-dimensional spatial point in the trajectory field, and the trajectory data is aligned with the jet velocity data through spatial location mapping. After data alignment, the angular change trend of the rebound path at each point can be extracted based on the time axis, and the rate of change of the rebound direction at that point can be calculated. Simultaneously, the amplitude of the surface shear displacement caused by the jet disturbance is integrally calculated to extract the shear displacement response amplitude. Then, the jet velocity disturbance at that point is frequency-domain transformed to obtain the corresponding spectrum curve, and the main energy feature region in the frequency domain is extracted to obtain the spectral disturbance energy feature value. Finally, using spatial coordinates as row indices and the three disturbance factors as column indices, a jet disturbance parameter fusion matrix is ​​constructed, thereby achieving spatial aggregation modeling of disturbance characteristics and forming a high-dimensional comprehensive expression of local disturbance behavior.

[0070] The perturbation amplitude and jet velocity perturbation spectrum data at corresponding locations refer to perturbation features from different data sources (visual trajectory and jet velocity sensing) under the same spatial coordinates, fused and aligned using a unified indexing method. The rate of change of rebound direction at each spatial location is based on the angular velocity of the rebound vector change at that point in consecutive frames, reflecting the instability of the jet particle rebound behavior. The shear displacement response amplitude represents the displacement intensity along the tangential direction of the concrete surface after being disturbed at that point, usually calculated by time integration of the surface micro-deformation field. The spectral perturbation energy characteristic value is obtained by performing a fast Fourier transform on the jet velocity signal, extracting the energy of the main frequency bands and calculating its integral value, used to measure the intensity of the perturbation at that point in the energy dimension. These perturbation features are collectively incorporated into the jet perturbation parameter fusion matrix, which can accurately present the distribution and energy structure of the jet perturbation field in the transition zone of the convex surface, and serves as the basic input for subsequent judgment of density reduction.

[0071] S203. Based on the weighted coupling relationship between the disturbance eigenvalues ​​at each position in the fusion matrix of spraying disturbance parameters, calculate the unit disturbance energy of each spatial point in the convex surface transition zone, and generate a disturbance energy density distribution map with continuous spatial coordinates as index, which is used to determine the change in reverse scouring pressure distribution in the convex surface transition zone during concrete spraying.

[0072] The reason for calculating the unit disturbance energy at each spatial point within the convex surface transition zone based on the weighted coupling relationship between the disturbance characteristic values ​​at each location in the spraying disturbance parameter fusion matrix, and generating a disturbance energy density distribution map using continuous spatial coordinates as an index, is that the convex surface transition zone exhibits significant spatial geometric abrupt changes, making it highly susceptible to becoming a point of superposition and accumulation of disturbance energy during concrete spraying. The concentration of reverse scour pressure is precisely caused by the nonlinear coupling evolution of these disturbance characteristics in space. By constructing a fusion matrix and extracting the coupling relationship, the interaction law of various disturbance factors in space, such as changes in rebound direction, shear displacement amplitude, and jet velocity disturbance, can be quantified and mapped into the unit disturbance energy at each spatial point. This calculation not only achieves accurate quantification of the impact of complex disturbances but also, after generating a disturbance energy density distribution map using continuous spatial coordinates as an index, comprehensively displays the flow direction, gradient changes, and abnormal accumulation of disturbance energy, thus providing data support and spectral basis for subsequent identification of abnormally concentrated areas of reverse scour pressure. This process is a crucial prerequisite for proactively predicting and locally identifying the potential weakening risk of the sprayed layer's density.

[0073] In this embodiment, S203 specifically refers to:

[0074] The rebound direction change rate, shear displacement response amplitude, and jet velocity disturbance spectrum intensity value of each spatial point are extracted from the jet disturbance parameter fusion matrix. A combined correlation table between each group of ternary disturbance feature values ​​is constructed. The local correlation coefficient is calculated by the convolution kernel sliding window method, and a position disturbance feature weight mapping map is generated accordingly.

[0075] To accurately identify the spatial variation trend of the disturbance response within the transition zone of the convex surface, three disturbance features for each spatial point need to be extracted from the jet disturbance parameter fusion matrix: the rate of change of the rebound direction, the amplitude of the shear displacement response, and the intensity of the jet velocity disturbance spectrum. For these features, a ternary disturbance feature combination is constructed, and a spatial index is established for each triplet. Then, a convolutional kernel window of a fixed size is set on the three-dimensional disturbance matrix, and the entire spatial region is scanned by sliding point-by-point to calculate the correlation between the local synergies of the ternary disturbance feature values ​​within the window. The calculation method can use Pearson correlation coefficient or covariance matrix to capture the coupling strength of the three types of disturbance features in local spatial cells. Finally, the correlation coefficients of each spatial point are mapped onto a spatial coordinate grid, forming a disturbance feature weight mapping map covering the entire transition zone of the convex surface. This map clearly reflects the inherent coupling relationship and the aggregation trend of disturbance intensity between different spatial point disturbance features.

[0076] Each set of ternary perturbation eigenvalues ​​refers to a vector composed of three physical parameters observed simultaneously at the same spatial point: the rate of change of the rebound direction, the amplitude of the shear displacement response, and the intensity of the jet velocity perturbation spectrum. This vector reflects the composite response behavior of the jet perturbation across different physical dimensions. The convolutional kernel sliding window approach is a computational strategy based on local region feature extraction. By moving a fixed-dimensional window within the three-dimensional perturbation matrix, the feature set within the current window is analyzed and calculated at each movement, capturing the local correlation and structural patterns of perturbation features in space. The perturbation feature weight mapping diagram is a graphical model formed by projecting the correlation coefficient of perturbation features calculated at each spatial point onto its three-dimensional spatial coordinates as weight values. It is used to express the spatial distribution pattern of perturbation coupling relationships and is the core foundational data model for subsequent identification of perturbation energy accumulation areas and density weakening risk areas.

[0077] Based on the weight coupling relationship of perturbation features at different locations in the location perturbation feature weight mapping map, a weighted integral operation is performed to calculate the unit perturbation energy corresponding to each spatial point. The unit perturbation energy distribution body is formed according to the three-dimensional coordinate order of the spatial points, and a spatial perturbation energy index model is constructed.

[0078] To further quantify the spatial distribution characteristics of spraying disturbance within the transition zone of a convex surface, it is necessary to perform weighted integral calculations based on the weighted coupling relationships between disturbance features at different spatial points in the positional disturbance feature weight mapping map. Specifically, for each discrete coordinate point in three-dimensional space, the rebound direction change rate, shear displacement response amplitude, and spray velocity disturbance spectrum intensity corresponding to that point are used as input factors. These are combined with the disturbance coupling weight coefficients calculated in the mapping map, and the disturbance response intensity at that point is solved using a weighted integral method to obtain the unit disturbance energy. The unit disturbance energy reflects the comprehensive disturbance intensity experienced by a single spatial location point during concrete spraying disturbance, expressed in joules per cubic meter or equivalent energy density. After calculating the unit disturbance energy for all spatial points, they are sequentially arranged according to the continuous topological structure of the three-dimensional coordinates to generate a unit disturbance energy distribution volume with a spatial topological index. Finally, this distribution volume is mapped into a spatial disturbance energy index model with spatial directionality and disturbance intensity grading attributes. This model is not only used to visualize the spatial distribution of disturbance energy, but also serves as the basic framework for subsequent identification of areas of concentrated backflow and potential areas of weakened compaction. The term "weighted coupling relationship" refers to the weighted distribution pattern determined through local correlation analysis between disturbance eigenvalues; "weighted integral operation" refers to the integration and superposition of disturbance factors using weighted values ​​to reflect their comprehensive impact; "unit disturbance energy" is a local quantitative expression of energy input during the jetting process; and the "spatial disturbance energy index model" is a spatial mapping of disturbance energy within a three-dimensional structure, used to support subsequent data analysis and control judgments.

[0079] Based on the spatial disturbance energy index model, the gradient flow direction and density accumulation degree of disturbance energy in space are calculated. Based on the abrupt change in the slope of disturbance energy change, the abnormal concentration area of ​​reverse scour pressure is identified, and a disturbance energy density distribution map is generated to determine the change in reverse scour pressure distribution in the convex curved surface transition area during concrete spraying.

[0080] When identifying back scour anomalies based on a spatial disturbance energy index model, it is necessary to first perform three-dimensional spatial continuous interpolation on the unit disturbance energy distribution to construct a continuous spatial disturbance energy function field. Based on this function field, spatial gradient analysis is used to calculate the rate and direction of change of disturbance energy in three-dimensional space, obtaining a gradient flow map to indicate the propagation path of the disturbance energy. Simultaneously, a density aggregation map is calculated by combining the concentration of unit disturbance energy in local areas to determine whether disturbance energy accumulates in certain structural depressions or transition zones. Next, first-order derivative slope analysis is performed on the gradient change map to identify locations where the rate of change of disturbance energy abruptly increases, i.e., abrupt change zones. These abrupt change zones typically represent locations where the stress state of the concrete sprayed layer changes drastically, i.e., anomaly points where back scour pressure is highly concentrated. By marking these abrupt change zones, sensitive areas that may be weakened in density due to shear disturbance and rebound scour can be accurately located. Combined with the disturbance energy value mapping, a disturbance energy density distribution map is generated, where each point corresponds to a set of energy flow direction and aggregation intensity indicators, forming a high-resolution spatial scour influence map. The term "gradient flow direction of continuously varying disturbance energy in space" refers to the vector trajectory of energy spreading in various directions in a three-dimensional structure; "density concentration" refers to the spatial concentration of local energy; "abrupt slope region of disturbance energy change" indicates a section where the disturbance intensity increases sharply with location, which often indicates an important precursor to potential structural weakening or damage risk.

[0081] S3. Link the disturbance energy density distribution map with the surface response curve of the initial setting stage of concrete to establish a density weakening prediction factor model for evaluating the integrity status of the sprayed layer, and output the weakening risk index.

[0082] In this embodiment, S3 specifically includes the following steps:

[0083] S301. Align the disturbance energy value of each spatial point in the disturbance energy density distribution map with the surface response curve collected at the corresponding spatial point during the initial setting stage of concrete, and extract the rate of change of disturbance energy change, response delay interval and response recovery time period in the response curve as linkage input variables.

[0084] When constructing the disturbance-response linkage input structure, it is necessary to align the disturbance energy value at each spatial point in the disturbance energy density distribution map with the surface response curve collected at the same spatial point during the initial setting stage of concrete. Specifically, this can be achieved by constructing a unified three-dimensional spatial coordinate index grid and mapping the energy point array of the disturbance energy density map with the sampling nodes of the response curve to achieve accurate registration of the two data sources. Based on this, the temporal variation trend of the disturbance energy value is extracted for each alignment point, and the peak acceleration / deceleration rate of the signal response in the corresponding response curve, the time period required for the response signal to first deviate from the initial steady state (i.e., the response delay interval), and the time required for the response signal to return to the steady state after the disturbance ends (i.e., the response recovery time period) are calculated. Through the above operations, a set of linkage input variables containing disturbance energy and response characteristic information is formed, which serves as the input structure for subsequent prediction modeling, comprehensively reflecting the sensitivity and self-stabilizing ability of the sprayed material to disturbance energy during the initial setting process.

[0085] The initial setting stage of concrete refers to the initial stage after the material is sprayed and begins to physically shrink and form its early structure. The sprayed structure in this stage is extremely sensitive to disturbance energy, representing a critical window period where density defects are most likely to occur. Changes in disturbance energy characterize the behavior of disturbance intensity over time relative to input pressure; the rate of change in response amplitude reflects the change in the slope of the concrete's deformation response to disturbance, used to assess the degree of stress release or accumulation; the response delay interval reflects the hysteresis response capability of the sprayed layer from disturbance triggering to the appearance of a measurable response, reflecting the integrity of its internal structure; and the response recovery time reflects the ability of the sprayed layer to quickly recover to a steady state after the disturbance ends, used to determine its self-healing and sealing characteristics. These indicators, when comprehensively extracted, constitute a complete set of disturbance-response linked input variables, helping to accurately capture potential signs of weakening sprayed layer density at the micromechanical behavior level.

[0086] S302. Based on the linked input variables, construct a disturbance response coupled dataset, use linear discriminant clustering and gradient regression screening algorithms to extract key parameter sets that reflect the characteristics of spray layer density weakening, establish a density weakening prediction factor model for evaluating the integrity status of the spray layer, and generate density weakening prediction factor values ​​for each spatial point.

[0087] S303. The density weakening prediction factor value is mapped in an orderly manner according to the spatial point coordinates, and the local gradient change trend of the disturbance energy density distribution map is superimposed to calculate the overall weakening degree of the spray layer and generate a weakening risk index, which is used to reflect the spatial distribution of the integrity state of the spray layer.

[0088] This method links disturbance response data with concrete surface response and assesses the integrity of the sprayed layer by constructing a density weakening prediction factor model and generating a weakening risk index. The primary goal is to address the difficulty in real-time identification and quantification of hidden density defects under complex construction conditions. By integrating disturbance energy density distribution maps with the initial setting stage response behavior of concrete, the actual feedback of spraying disturbances on the microscopic damage of the structural layer can be captured. Supported by linear discriminant clustering and gradient regression, key variable groups influencing density weakening are further extracted, and a density weakening prediction factor model with discriminative capabilities is constructed, effectively avoiding judgment bias caused by redundant dependent variables. Through spatial point mapping and local disturbance gradient superposition, the weakening degree of the sprayed layer in different regions is comprehensively calculated, and a weakening risk index is output. This enables the spatial visualization and identification of the sprayed layer's density state, providing a quantitative basis for dynamically adjusting spraying process parameters and intervening in the formation of potential weak points in advance. This mechanism achieves closed-loop modeling from disturbance excitation to structural response to risk assessment, breaking through the limitations of traditional reliance on static specifications and empirical judgments, and possesses high data-driven and structural adaptability.

[0089] In this embodiment, S302 specifically refers to:

[0090] A perturbation response coupled dataset is constructed using the perturbation energy density distribution map and the perturbation energy value, response amplitude change rate, response delay interval and response recovery time period at the corresponding spatial location in the surface response curve as input, and the input variables are bound to the spray layer location using spatial coordinates as index;

[0091] The key to constructing a disturbance response coupled dataset lies in fusing data to establish a one-to-one correspondence between the sprayed disturbance energy and the response characteristics of concrete materials during the initial setting stage. Specifically, this involves first extracting the disturbance energy values ​​at each spatial location point from the disturbance energy density distribution map, and simultaneously extracting the surface response curve parameters obtained at the same spatial location using high-frequency response measurement technology, including the rate of change of response amplitude, response delay interval, and response recovery time. Subsequently, a unified three-dimensional spatial coordinate system is used as the primary index to bind all variable values ​​to coordinates, thereby forming a multi-dimensional joint data table of disturbance response oriented towards the sprayed layer structure space. This data table constitutes the disturbance response coupled dataset, where each data record represents the correspondence between the external disturbance intensity and the material response characteristics at a specific spatial point within the sprayed layer. In implementation, this can be achieved by programming to construct a spatial hash table or a three-dimensional matrix index structure for data organization, and using data fusion tools for variable association and matching. The core value of this dataset construction method lies in providing a highly structured data foundation for subsequent pattern recognition, clustering judgment, and predictive modeling, ensuring a clear physical and spatial correlation between the impact of disturbances and the structural state, thereby improving the accuracy and reliability of identification when the density of the spray layer is weakened.

[0092] Linear discriminant clustering was performed on the perturbation-response coupling dataset to extract data clusters that exhibited the coupling behavior of abrupt changes in perturbation peaks and abnormal responses. Then, a gradient regression screening algorithm was used to remove weak interfering variables and select key parameter groups that reflect the weakening state of the spray layer's compactness.

[0093] The purpose of linear discriminant clustering on the disturbance-response coupled dataset is to identify spatial locations where the disturbance energy changes drastically and the concrete response is abnormal. Specifically, based on the disturbance energy value and response curve parameters recorded at each spatial point, linear discriminant analysis (LDA) is used to reduce the dimensionality of the data, making the distribution boundaries of different types of responses clearer. Then, clustering algorithms such as K-means or density clustering are used to divide the data into several clusters. Among these clusters, regions where the disturbance energy exhibits an instantaneous peak and the corresponding response parameters show hysteresis, backhysteresis, or extension are particularly important; these clusters are defined as data clusters exhibiting the coupling behavior of abrupt changes in disturbance peaks and abnormal responses. Subsequently, for the data within these clusters, a gradient regression screening algorithm is used to perform multiple rounds of regression analysis on input variables such as disturbance energy value, rate of change of response amplitude, delay interval, and recovery time period. This evaluates the predictive ability of each variable for abnormal sprayed layer response and removes weak variables with low regression contributions. This process ultimately identifies a set of key parameters that can efficiently characterize risk areas where the spray layer may experience density reduction under external jetting disturbances. These parameters form the core input variables for establishing subsequent density reduction prediction factor models. The entire process considers data structuring, statistical significance assessment, and engineering interpretability to ensure that the model possesses physical rationality and accurate predictive capabilities.

[0094] The key parameter set is input into the multi-factor comprehensive evaluation framework. The peak sensitivity of the disturbance, the hysteresis degree of the response curve and the spatial consistency are used as evaluation indicators to construct a compaction weakening prediction factor model. The compaction weakening prediction factor value is calculated for each spatial point for spray layer integrity status assessment.

[0095] The purpose of inputting key parameter sets into a multi-factor comprehensive evaluation framework is to establish a mathematical model that can accurately reflect the weakening trend of sprayed layer density. This evaluation framework quantitatively assesses the perturbation-response characteristics of each spatial point by simultaneously introducing indicators from multiple dimensions. In specific implementation, the key parameter sets selected from the perturbation-response coupling dataset are first input into a weighted comprehensive scoring model. The three core evaluation indicators in the model include peak perturbation sensitivity, response curve hysteresis, and spatial consistency. Peak perturbation sensitivity refers to the response intensity of each spatial point to changes in the initial setting response of concrete, reflecting the directness of the perturbation's impact on the dense structure. Response curve hysteresis is used to capture micro-defects or energy dissipation characteristics within the structure by analyzing the asymmetry between surface response delay and recovery. Spatial consistency measures the homogeneous stability of the sprayed layer structure by evaluating the consistency of the changing trends of perturbation and response between adjacent spatial points. Based on this, multi-factor weighted scoring calculations are performed on each spatial point to generate a density weakening prediction factor model. The density weakening prediction factor value output by this model is used to describe the warning level of the sprayed layer integrity status at that point. For example, when the disturbance peak sensitivity is high, the hysteresis is large, and the spatial consistency is low, this point will be identified as a potential weakening area. This method can identify hidden weak areas inside the sprayed layer in advance, providing a basis for subsequent construction control and improving the control accuracy and reliability of impermeability performance.

[0096] In this embodiment, S303 specifically refers to:

[0097] The compactness reduction prediction factor values ​​are arranged in an orderly manner according to the three-dimensional coordinates of spatial points, and a compactness reduction state field is constructed under a unified coordinate system to realize the continuous mapping of the reduction factor in space.

[0098] Arranging the density weakening prediction factor values ​​in an ordered manner according to the three-dimensional coordinates of spatial points and constructing a density weakening state field under a unified coordinate system is to transform discrete prediction data into a spatially continuous three-dimensional field, enabling a comprehensive analysis of the integrity state of the sprayed structure at different spatial locations. Specifically, firstly, the density weakening prediction factor value of each spatial point is bound to its corresponding three-dimensional coordinate, using a unified reference coordinate system to ensure all data are mapped within the same spatial dimension. Subsequently, these discrete factor values ​​are continuously completed in three-dimensional space using a three-dimensional mesh interpolation algorithm or spline function fitting technique, constructing the density weakening state field. This density weakening state field is essentially a three-dimensional tensor indexed by spatial coordinates and with factor values ​​as amplitudes, accurately depicting the density weakening intensity distribution of the sprayed structure in various spatial regions. For example, during spraying construction, areas with significantly increased factor values ​​in the state field can be observed in real time, quickly locating potential weakening points. The advantage of this approach is that it can expand point-based judgment results into area-based or even volume-based risk assessment scenarios, providing a higher-dimensional information foundation for spray layer integrity analysis and a clear reference for the spatial deployment of subsequent control strategies.

[0099] Based on the compactness weakening state field, the local gradient change trend of the corresponding spatial point in the perturbation energy density distribution map is used as an incremental weight factor, and it is superimposed and fused with the compactness weakening prediction factor value point by point to construct a multi-dimensional perturbation-compactness comprehensive evaluation matrix.

[0100] Based on the compaction weakening state field, the local gradient change trend of corresponding spatial points in the perturbation energy density distribution map is used as an incremental weighting factor and fused point-by-point with the compaction weakening prediction factor value to achieve a dual-coupled evaluation of perturbation intensity and compaction state. Specifically, the local gradient change trend is first obtained by calculating the energy change rate between each spatial point and its neighboring points in the perturbation energy density distribution map. This reflects the concentration and slope of the jet perturbation in the local space. For example, if the perturbation energy in a certain area shows a sharp increase in space, it indicates that there may be jet backlash accumulation, and a higher evaluation weight should be given. Subsequently, these local gradient change trends are mapped to the corresponding spatial points as incremental weighting factors and fused with the compaction weakening prediction factor value in the compaction weakening state field to form a comprehensive evaluation matrix containing both perturbation and compaction data. This matrix can simultaneously reflect the spatial characteristics of the jet layer's response intensity and structural integrity under perturbation. For example, at a given location, if the perturbation gradient is sharp and the density reduction index is high, then that point will have a larger weight in the comprehensive matrix and be more significant for assessment and warning. The multidimensional perturbation-density comprehensive evaluation matrix constructed in this way can serve as the basic data support for subsequent overall risk index generation and regional intervention judgment.

[0101] Local integration calculations are performed on the multidimensional perturbation-density comprehensive evaluation matrix. The overall degree of spray layer weakening is calculated by combining the spatial point clustering trend. Based on the set grading standard, a weakening risk index is output to characterize the spatial distribution characteristics of the spray layer integrity state.

[0102] Local integration calculations are performed on the multidimensional perturbation-density comprehensive evaluation matrix to summarize and extract trends of the local perturbation intensity and density reduction degree at each spatial point within a spatial range. Specifically, a local integration operation is first performed on the evaluation matrix in a sliding window within a three-dimensional coordinate system to calculate the superposition intensity of perturbation energy and density weakening factors in each region, generating corresponding local cumulative values. Subsequently, by analyzing the aggregation trend of these local cumulative values ​​across the entire spatial distribution, high-density regions and continuous change patterns of weakening features are extracted, thereby constructing a spatial mapping map of the overall spray layer weakening degree. Regarding the setting of grading standards, a mapping relationship between different perturbation accumulation levels and spray layer instability probabilities is established using pre-experimental data, and multiple risk level intervals are defined, such as mild weakening, moderate weakening, and severe weakening. Based on this, a weakening risk index corresponding to each sub-region is output according to the cumulative intensity distribution of the spray layer region. For example, if multiple spatial points within a region show high coupling values ​​in the local integration calculation and are continuously distributed in a high gradient region, then that region is marked as high-risk in the weakening risk index. This method not only enables precise identification of the integrity status of the sprayed layer, but also provides a basis for parameter adjustment in advance during construction.

[0103] S4. By combining the risk reduction index with the jet trajectory time series data and the jet beam effect distribution results, nonlinear fitting is performed to determine whether the density of the formed spray layer in the convex curved surface transition zone is damaged, and the spray layer structure state is divided into multiple levels.

[0104] In this embodiment, S4 specifically refers to:

[0105] The index value of each spatial point in the weakening risk index is aligned one-to-one with the trajectory displacement curve of the corresponding spatial point in the jet trajectory time series data, as well as the jet intensity and jet duration of the corresponding spatial point in the jet effect distribution results according to the three-dimensional spatial coordinates. A nonlinear fitting input dataset indexed by three-dimensional spatial coordinates is constructed, and amplitude standardization is performed on all input variables to ensure the stability of the fitting operation.

[0106] In concrete shotcreting, to accurately assess the nonlinear relationship between changes in spray layer density and spraying parameters, it is necessary to uniformly index and standardize multiple spatiotemporal distribution variables acquired during the spraying process. First, for the mitigation risk index, spray trajectory time series data, and spray beam distribution results, the position of each spatial point in the three-dimensional coordinate system is used as a unique index benchmark to ensure point-by-point alignment of data from different sources under the same spatial semantics. The spray trajectory time series data provides the trajectory offset information of the spray beam particles acting on the surface at each time frame, while the spray beam distribution results record the intensity of the spray beam action on each spatial point and the duration of that intensity at that point. Through a time matching mechanism and spatial coordinate consistency mapping, the input variables corresponding to each three-dimensional point, such as the "mitigation risk index value, trajectory displacement curve, spray beam action intensity, and spray beam duration," are extracted to construct a nonlinear fitting input dataset.

[0107] After constructing the input dataset, all variables need to be standardized to avoid fitting bias caused by different units and numerical ranges. Specific methods include range normalization or Z-score normalization for each type of input variable, bringing their mean close to zero and variance to one, thereby enhancing the model's response consistency across different scales of data. For example, jet intensity might be measured in kilopascals, while trajectory displacement is in millimeters. Without amplitude standardization, the fitting algorithm might misjudge the importance of variables, affecting the reliability of the results. The nonlinear fitting input dataset is a high-dimensional input structure that integrates spatial point perturbation information with the relationship between jet parameter changes, providing necessary collaborative information support for the subsequent fitting model to identify the density weakening trend. In this process, "jet intensity" refers to the numerical distribution of particle impact force per unit area per unit time, "jet duration" refers to the cumulative time a particle maintains its action at that point, and the "nonlinear fitting input dataset" is a data structure containing standardized multidimensional variables, organized in an ordered manner according to three-dimensional coordinates.

[0108] Based on the nonlinear fitting input dataset, a multidimensional nonlinear fitting algorithm is used to establish a response fitting model between the risk reduction index and the jet trajectory time series data and the jet effect distribution results, and the fitting residual value and response offset of each spatial point are calculated. When the fitting residual value of a spatial point exceeds the fitting residual threshold and its response offset shows a continuous increasing trend, it is determined that the compactness of the formed spray layer in the convex curved surface transition area corresponding to the spatial point is damaged, and the spatial point is added to the set of compactness damage points.

[0109] When identifying the densification failure of shotcrete layers, it is necessary to establish a model that can describe the complex nonlinear relationship between the perturbation parameters of the spraying process and the response of the shotcrete layer. To achieve this goal, based on the previously constructed nonlinear fitting input dataset, a multidimensional nonlinear fitting algorithm is used to model the response relationship between the weakening risk index and the time series data of the spray trajectory and the distribution of the spray jet effect. This fitting algorithm can use methods such as support vector regression, neural network fitting, or higher-order polynomial regression to capture the complex nonlinear coupling characteristics between different variables. After establishing the response fitting model, the fitting output value of each spatial point is compared with the actual observed value to obtain the fitting residual value of that point. The residual value reflects the degree of deviation between the model prediction result and the actual perturbation response. At the same time, the trend of prediction error in the continuous time series is analyzed to extract the change curve of its response offset. If the fitting residual value of a certain spatial point is continuously higher than the set fitting residual threshold, and its response offset shows a continuous increasing trend in the time dimension, it means that the response characteristics of the shotcrete layer at that point have deviated from the normal perturbation model response range, and thus it can be determined that the densification of the shotcrete layer has been structurally weakened in this area.

[0110] Throughout the process, the "multidimensional nonlinear fitting algorithm" refers to a nonlinear function approximation method used for high-dimensional input variables to construct a response mapping relationship between perturbation input and mitigation risk. The "response fitting model" is a predictive framework obtained after algorithm training, used to deduce the theoretical mitigation risk index value based on the input variables. The "fitting residual value" is the difference between the actual observed value and the model's predicted value, used to measure the model's fitting accuracy at local points. The "fitting residual threshold" is an upper limit of error set through historical data statistics or cross-validation; exceeding this threshold indicates model fitting failure. A "persistent expansion trend" refers to an increasing absolute value of the residual or response offset on the time axis, indicating that the perturbation response at that point continuously deviates from the model's expectations, a significant indicator of structural performance degradation. Finally, this mechanism is used to screen all spatial points that meet the damage characteristics and collect them to form a dense damage point set, providing data support for spray layer damage assessment.

[0111] Based on the fitting residual amplitude, response offset, and degree of aggregation of spatial points in the set of dense failure points, a density grading standard is set, and the structural state of the formed sprayed layer in the convex curved surface transition zone is divided into multiple levels: dense and stable, slightly weakened, moderately weakened, and severely weakened, and a spatial level distribution map of the sprayed layer structural state is formed.

[0112] To achieve refined hierarchical management of the formed sprayed layer structure state in the transition zone of convex curved surfaces, a comprehensive judgment based on the quantitative characteristics of each spatial point in the set of dense failure points is required. Specifically, the following steps are taken: First, the corresponding fitting residual amplitude and response offset are extracted for each spatial point as the main indicators for measuring the degree of disturbance anomaly and response deviation. Simultaneously, the local spatial point density within the set of dense failure points is statistically analyzed, i.e., the degree of spatial point clustering, to determine whether the failure exhibits a regional concentration trend. Based on this, a denseness grading standard is defined through cluster analysis and an expert system. This standard can set multiple threshold intervals, corresponding to four levels: dense and stable, slightly weakened, moderately weakened, and severely weakened. Each spatial point is classified into the corresponding level based on the combination of its three parameters, thus completing the quantitative grading of the sprayed layer structure state.

[0113] The "Denseness Grading Standard" is a set of classification rules based on quantitative data clustering and engineering experience. It compresses multidimensional disturbance and response characteristics into structural density levels. "Dividing the structural state of the formed sprayed layer in the convex curved surface transition zone into multiple levels—dense and stable, slightly weakened, moderately weakened, and severely weakened" means that the density of the sprayed layer is no longer judged by a single binary standard of good and bad, but rather by a multi-level, refined evaluation based on the coupled response of disturbances. This facilitates the formulation of subsequent differentiated reinforcement or repair measures. The "Spatial Distribution Map of Sprayed Layer Structural State" is a graphical representation that visualizes the level states corresponding to all spatial points in a three-dimensional spatial coordinate system. It uses color, density, or morphological coding to show the structural health level of different areas of the sprayed layer, making the assessment results intuitive, hierarchical, and valuable for decision support. This process not only improves the resolution of sprayed layer integrity diagnosis but also provides a data foundation for dynamic monitoring and risk warning.

[0114] S5. Based on the classification of the spray layer structure, the spray energy density, spray angle, path speed and mixture ratio are adjusted in a coordinated manner, and the disturbance energy density distribution map is updated in real time to correct the density reduction prediction factor model, so as to realize the dynamic closed-loop control of the spray layer density.

[0115] In this embodiment, S5 specifically refers to:

[0116] Based on the division of the spray layer structure state, a parameter control mapping relationship is established. Different density level regions are matched with spray energy density, spray angle, path speed and mixture ratio parameters to generate a spray control target set and bind it to the structural state space coordinates to indicate the process adjustment strategies required for different spatial points.

[0117] To achieve targeted spray control, a parameter control mapping relationship needs to be constructed based on the classification of the spray layer structure. This process can be achieved by setting predefined correspondence rules between density levels and spray parameters. For example, dense and stable regions can be associated with lower spray energy density and conventional proportions, slightly weakened regions with medium spray energy density and faster path speed, moderately weakened regions with increased spray angle and spray intensity, and severely weakened regions with simultaneously increased spray energy density, reduced path speed, and adjusted proportions of accelerators and binders in the mixture. Each density level corresponds to a complete set of spray parameter combinations, which are then bound to specific spatial points in a three-dimensional coordinate system, thus forming a spray control target set. This process can be achieved by constructing a spatial attribute table, using coordinate points, density levels, and corresponding spray parameter combinations as record items, and automatically distributing parameters by calling this table during system operation.

[0118] Once the injection control target set is generated, it can be used to guide the injection process strategy at different spatial points, ensuring that the injection behavior is precisely adapted to the spray layer state. Specifically, the injection energy density is controlled by adjusting the injection pump pressure and nozzle flow rate; the injection angle is dynamically adjusted by a multi-axis linkage nozzle mechanism; the path speed is set by a nozzle motion controller according to a coordinate sequence; and the mixture ratio is automatically adjusted by an intelligent mixing unit based on real-time signals. Changes in the density level at each spatial point will trigger adjustments to the corresponding injection parameters in real time, thereby achieving differentiated configuration of process parameters in the spatial dimension. This approach ensures efficient allocation of injection resources and precise control of density recovery, forming a crucial foundation for achieving dynamic closed-loop control of spray layer density.

[0119] Based on the injection energy density output module, nozzle angle control module, path motion control module and mixing ratio supply module of the injection control target set linkage control injection system, the system collects the disturbance response data generated during the injection process in real time while performing differentiated injection operations, and calculates the disturbance energy parameters in the new round of injection intervention area online.

[0120] When controlling the density of the spray coating, it is necessary to rely on the parameter configuration corresponding to each spatial point in the spray control target set to link and control the various key execution modules within the spraying system. The spraying energy density output module adjusts the discharge pressure and matches the nozzle diameter through a high-pressure delivery pump to control the kinetic energy density of the projected material per unit time; the nozzle angle control module adjusts the angle between the nozzle and the target surface based on a multi-axis servo drive structure to achieve spatial variation of the spray beam direction; the path motion control module achieves precise control of the continuous spraying trajectory by programming the displacement path and moving speed of the nozzle; the mixing ratio supply module consists of a weighing sensor system and a quantitative feeding device, which automatically adjusts the material ratio by controlling the water-cement ratio, aggregate ratio, and additive injection rate. These four modules jointly respond to the input parameters in the spray control target set to realize differentiated spraying operation strategies between regions, ensuring that the spray coating response accurately corresponds to the structural state.

[0121] During the differentiated spraying operation, it is also necessary to synchronously collect disturbance response data in the new round of spraying intervention area. By deploying high-frequency imaging sensors and surface stress sensing devices in the vicinity of the nozzle, the rebound trajectory of concrete particles impacting the surface, surface shear disturbance, and displacement feedback are recorded in real time to construct disturbance response time-series data. Subsequently, the sliding window algorithm and real-time calculation module can be used at the data acquisition terminal to calculate the current disturbance energy value of each spatial point online, including the spray jet disturbance frequency response, disturbance amplitude variation trend, and disturbance persistence parameters, and regenerate the local disturbance energy density map. This process provides a real-time feedback channel for the iterative correction of the subsequent prediction factor model, constituting a key supporting link in the closed-loop control mechanism of spray layer density.

[0122] The updated disturbance energy parameters in the injection intervention area are fused with the original disturbance energy density distribution map to construct a dynamic update map of disturbance energy density. The dynamic update map is then used as input to adjust the weights of characteristic variables in the density reduction prediction factor model in real time, forming a linkage mapping model between disturbance control and density assessment, thereby achieving dynamic closed-loop control of the spray layer density.

[0123] During the injection intervention process, to maintain a continuous response relationship between the spray layer compactness assessment and the injection disturbance behavior, it is necessary to fuse the newly acquired disturbance energy parameters within the injection intervention area with the previously constructed disturbance energy density distribution map. This process can be achieved by mapping the newly acquired disturbance energy values ​​to a unified three-dimensional coordinate system using a spatial point correspondence method. A weighted overlay algorithm combined with a time window moving average mechanism is then used to calculate the dynamic change value of the disturbance energy at each spatial point, thereby constructing a dynamically updated disturbance energy density map. This fusion process not only improves the temporal sensitivity and spatial accuracy of the disturbance energy data but also provides higher-resolution input information for subsequent model correction.

[0124] Based on the data variation trend in the dynamic update map of perturbation energy density, the weight distribution of each feature variable in the compactness weakening prediction factor model can be corrected in real time. By using the dynamic update map as input, combined with changes in perturbation peak value, shear response amplitude adjustment, and spatial energy gradient changes, a multi-factor weighted update algorithm is used to adjust the perturbation feature weights in the original model point by point, generating a linked mapping model covering the global space. Under this model, perturbation control behavior and compactness state judgment form a closed-loop response. Each round of injection adjustment triggers automatic model updates and is reflected in the generation of injection control parameters for the next round, thereby achieving continuous feedback and dynamic closed-loop control of the spray layer compactness. This approach ensures that the model's adaptability is highly consistent with the execution logic of the injection system, which is a key means to achieve high-precision spray layer quality management.

[0125] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means (e.g., infrared, wireless, microwave, etc.). A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0126] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0127] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0128] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0129] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0130] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0131] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A high-density shotcrete impervious control method for a special-shaped roof steel structure floor support plate, characterized in that, Specifically comprising the following steps: S1, the three-dimensional structure of the special-shaped roof steel structure floor support plate is unfolded and analyzed, the positive gradient sudden increase boundary in the spatial continuous curvature change is extracted, the convex surface turning area is identified, and the curvature mutation zone is marked in the convex surface turning area as the subsequent monitoring area; S2, the rebound trajectory field and the shear disturbance trajectory field are synchronously collected in the marked convex surface turning area during the concrete spraying process, the reverse scouring pressure distribution change of the convex surface turning area in the concrete spraying process is determined combined with the jet velocity disturbance frequency spectrum data, and the disturbance energy density distribution graph is formed; S3, the disturbance energy density distribution graph and the surface response curve in the initial setting stage of the concrete are linked to calculate, a compactness weakening prediction factor model for evaluating the integrity state of the sprayed layer is established, and a weakening risk index is output; S4, the weakening risk index is combined with the spraying trajectory time sequence data and the jet beam action distribution result to perform nonlinear fitting, whether the compactness of the formed sprayed layer in the convex surface turning area is damaged is judged, and the sprayed layer structure state is divided into multiple levels; S5, according to the division result of the sprayed layer structure state, the spraying energy density, the spraying angle, the path speed and the mixture material ratio are linked to control, and the disturbance energy density distribution graph is updated in real time to modify the compactness weakening prediction factor model, so that the dynamic closed-loop control of the compactness of the sprayed layer is realized.

2. The high-density shotcrete imperviousness control method for the special-shaped roof steel structural floor support plate according to claim 1, characterized in that, S1 specifically is: A three-dimensional structure model containing node spatial coordinates and node normal vectors is constructed for the special-shaped roof steel structure floor support plate, the curvature sequence of the node normal vectors is calculated, the three-dimensional structure unfolding analysis of the special-shaped roof steel structure floor support plate is completed, and the spatial continuous curvature change distribution is obtained; In the spatial continuous curvature change distribution, the positive gradient sudden increase boundary is extracted by calculating the positive gradient of the curvature with the spatial position, and the curvature sudden increase boundary set is established at the position where the positive gradient suddenly increases, which is used to identify the candidate convex surface area with sudden increase characteristics in the spatial continuous curvature change; In the curvature sudden increase boundary set, the convex surface turning area is identified according to the rotation aggregation trend of the node normal vector and the continuity of the curvature growth direction, and the curvature mutation zone is marked with the convex surface turning area as the center and outwardly extended to the curvature change stable area, so as to complete the monitoring area establishment of the convex surface turning area.

3. The high-density shotcrete imperviousness control method for the special-shaped roof steel structural floor support plate according to claim 1, characterized in that, S2 specifically includes the following steps: S201, during the concrete spraying process, a synchronous imaging collection array is arranged in the marked convex surface turning area, which is used to continuously record the rebound path formed after the concrete particles impact and the surface tangential displacement caused by the spraying disturbance, and the rebound trajectory field and the shear disturbance trajectory field are obtained by reconstruction through a unified time axis; S202, the disturbance amplitude of the corresponding position points in the rebound trajectory field and the shear disturbance trajectory field is synchronously aligned point by point with the jet velocity disturbance frequency spectrum data, the rebound direction change rate, the shear displacement response amplitude and the spectral disturbance energy characteristic value at each spatial position are extracted, and a spraying disturbance parameter fusion matrix is constructed; S203, according to the weight coupling relationship between the position disturbance characteristic values in the jet disturbance parameter fusion matrix, the unit disturbance energy of each space point in the convex surface turning area is calculated, and a disturbance energy density distribution map is generated with the spatial continuous coordinates as the index, which is used to determine the reverse scouring pressure distribution change of the convex surface turning area in the concrete jetting process.

4. The high-density shotcrete imperviousness control method for the special-shaped roof steel structural floor support plate according to claim 3, characterized in that, S203 specifically includes the following steps: The rebound direction change rate value, the shear displacement response amplitude value and the jet beam speed disturbance frequency spectrum intensity value of each space point are extracted from the jet disturbance parameter fusion matrix, and a combination correlation table is constructed among each group of three disturbance characteristic values. The local correlation coefficient is calculated by means of convolution kernel sliding window, and a position disturbance characteristic weight mapping diagram is generated accordingly; According to the weight coupling relationship of different position disturbance characteristics in the position disturbance characteristic weight mapping diagram, weighted integral operation is performed to calculate the unit disturbance energy corresponding to each space point, and a unit disturbance energy distribution body is formed according to the order of three-dimensional coordinates of the space point, and a spatial disturbance energy index model is constructed; Based on the spatial disturbance energy index model, the gradient flow direction and the density aggregation degree of the continuous change of the disturbance energy in space are calculated, and the reverse scouring pressure abnormal concentration area is identified according to the disturbance energy change slope mutation area, and a disturbance energy density distribution map is generated, which is used to determine the reverse scouring pressure distribution change of the convex surface turning area in the concrete jetting process.

5. The high-density shotcrete impervious control method for the special-shaped roof steel structural floor support plate according to claim 1, characterized in that, S3 specifically includes the following steps: S301, the disturbance energy value of each space point in the disturbance energy density distribution map is aligned with the surface response curve collected at the corresponding space point in the initial setting stage of the concrete, and the disturbance energy change and the response amplitude change rate, the response delay interval and the response recovery time period in the response curve are extracted as linkage input variables; S302, based on the linkage input variables, a disturbance response coupling data set is constructed, a linear discriminant clustering and gradient regression screening algorithm is used to extract the key parameter group reflecting the densification weakening feature of the sprayed layer, a densification weakening pre-judgment factor model for evaluating the integrity state of the sprayed layer is established, and a densification weakening pre-judgment factor value is generated for each space point; S303, the densification weakening pre-judgment factor value is sequentially mapped according to the space point coordinates, the local gradient change trend of the disturbance energy density distribution map is superimposed, the overall weakening degree of the sprayed layer is calculated, and a weakening risk index is generated, which is used to reflect the spatial distribution of the sprayed layer integrity state.

6. The high-density shotcrete imperviousness control method for the special-shaped roof steel structural floor support plate according to claim 5, characterized in that, S302 specifically includes the following steps: The disturbance energy value, the response amplitude change rate, the response delay interval and the response recovery time period of the corresponding space position in the disturbance energy density distribution map and the surface response curve are taken as inputs to construct a disturbance response coupling data set, and the input variables are bound to the sprayed layer position with the spatial coordinates as the index; The linear discriminant clustering operation is performed on the disturbance response coupling data set, the data clusters showing the disturbance peak mutation and response abnormal coupling behavior are extracted, and then the gradient regression screening algorithm is used to remove the interference weak variables, and the key parameter group reflecting the densification weakening state of the sprayed layer is screened out; The key parameter groups are input into a multi-factor comprehensive evaluation framework, peak sensitivity of perturbation, response curve hysteresis degree and spatial consistency are taken as evaluation indexes, a compactness weakening prediction factor model is constructed, and a compactness weakening prediction factor value is calculated for each spatial point for spray layer integrity state evaluation.

7. The high-density shotcrete imperviousness control method for the special-shaped roof steel structural floor support plate according to claim 5, characterized in that, S303 specifically is: The compactness weakening prediction factor values are sequentially arranged according to the three-dimensional coordinates of the spatial points, and a compactness weakening state field is constructed in a unified coordinate system to realize continuous mapping of the weakening factor in space; Based on the compactness weakening state field, the local gradient change trend of the corresponding spatial points in the perturbation energy density distribution diagram is taken as an incremental weight factor, which is point-by-point superimposed and fused with the compactness weakening prediction factor value to construct a multi-dimensional perturbation-compactness comprehensive evaluation matrix; The multi-dimensional perturbation-compactness comprehensive evaluation matrix is subjected to local integral calculation, the overall spray layer weakening degree is calculated in combination with the spatial point aggregation trend, and the weakening risk index is output according to the set classification standard to represent the spatial distribution characteristics of the spray layer integrity state.

8. The high-density shotcrete impervious control method for the special-shaped roof steel structural floor support plate according to claim 1, characterized in that, S4 specifically is: The index value of each spatial point in the weakening risk index is aligned with the trajectory displacement curve of the corresponding spatial point in the jet trajectory time series data, and the jet beam action intensity and duration of the corresponding spatial point in the jet beam action distribution result according to the three-dimensional spatial coordinates to construct a nonlinear fitting input data set indexed by three-dimensional spatial coordinates, and perform amplitude normalization processing on all input variables to ensure the stability of the fitting operation; Based on the nonlinear fitting input data set, a response fitting model between the weakening risk index and the jet trajectory time series data and the jet beam action distribution result is established by using a multi-dimensional nonlinear fitting algorithm, and the fitting residual value and response offset of each spatial point are calculated; when the fitting residual value of the spatial point exceeds the fitting residual threshold and its response offset shows a persistent expansion trend, it is determined that the compactness of the formed spray layer in the convex surface turning area corresponding to the spatial point is damaged, and the spatial point is added to the compactness damage point set; According to the fitting residual amplitude, response offset and aggregation degree of the spatial points in the compactness damage point set, the compactness classification standard is set to divide the structure state of the formed spray layer in the convex surface turning area into compactness stable, slight weakening, moderate weakening and severe weakening, and a spatial grade distribution map of the spray layer structure state is formed.

9. The high-density shotcrete imperviousness control method for the special-shaped roof steel structural floor support plate according to claim 1, characterized in that, S5 specifically is: According to the division result of the spray layer structure state, a parameter control mapping relationship is established, different compactness level regions are matched with jet energy density, jet angle, path speed and mixture proportion parameters to generate a jet control target set and bind it to the structure state spatial coordinates to indicate the process adjustment strategy required by different spatial points; Based on the jet control target set, the jet energy density output module, the jet angle control module, the path motion control module and the mixed material proportion supply module of the jet system are jointly controlled to perform differential jet operation while collecting the disturbance response data generated during the jet process in real time, and the disturbance energy parameters in the new round of jet intervention area are calculated online; The updated disturbance energy parameter of the spray intervention area is fused with the original disturbance energy density distribution map, a disturbance energy density dynamic update map is constructed, and the feature variable weight in the compactness weakening prediction factor model is corrected in real time with the dynamic update map as input, forming a linkage mapping model between disturbance regulation and compactness evaluation, and realizing dynamic closed-loop regulation of the compactness of the spray layer.

Images