River state evaluation system for arch bridge water cutting construction

Abstract

The application discloses a river state evaluation system for arch bridge water demolition and cutting construction, and relates to the technical field of water bridge construction. Firstly, arch bridge demolition and cutting construction is performed to obtain river state data. The river state data is processed to obtain first vector features and second vector features, and vector feature splicing is performed to obtain a river feature matrix. The river feature matrix is processed by using a clustering algorithm to obtain anchoring bearing evaluation data. A thermal map is rendered according to the meta model, a hierarchical alarm is triggered, and a construction decision instruction is output. The abnormal mode of the water surface environment state of the construction bridge demolition can be automatically identified, and adaptive adjustment is facilitated, thereby providing safety information support for arch bridge water demolition and cutting construction.

Description

Technical Field

[0001] This invention relates to the field of water bridge construction technology, and in particular to a river condition assessment system for the dismantling and cutting of arch bridges on water. Background Technology

[0002] As an important type of bridge spanning rivers, the demolition of arch bridges faces multiple challenges, including complex structural stresses and variable aquatic working environments. When using floating pontoons as construction platforms for the underwater cutting and demolition of arch bridges, the pontoons maintain their positioning and load-bearing stability through an anchoring system and temporary moorings. The safety of the entire construction system is highly dependent on the load-bearing capacity of the riverbed anchoring points and the real-time state of the river's hydrodynamic environment.

[0003] Traditional methods often employ point-based current meters or manual depth sounders for discrete, single-point measurements, failing to simultaneously acquire the flow field distribution, dynamic evolution of riverbed topography, seabed characteristics, and mechanical response of the floating vessel-anchor chain system across the entire construction area. Existing early warning systems typically issue single-parameter alarms based on preset fixed thresholds, unable to identify gradual deterioration of anchoring safety under the coupling effects of multiple factors, let alone detect unknown new risk patterns. For example, under certain conditions, even if the flow velocity and scour depth do not exceed limits, the coupling effect of uneven seabed density and flow field vibration can still lead to bollard tension imbalance and gradual anchor drift. Such complex risks are difficult to detect in a timely manner using traditional methods. Furthermore, the failure to quantify safety assessment results into actionable construction process adjustment parameters makes it difficult to provide specific command-level guidance for operations such as floating vessel ballast compensation, bollard tension correction, and cutting rate adjustment, resulting in a significant human decision-making delay between risk detection and control measures.

[0004] Therefore, there is an urgent need for a method that can perform multi-dimensional in-situ perception of the river channel conditions during the dismantling and cutting of arch bridges, intelligently cluster and evaluate them, and automatically generate construction decision instructions, so as to improve the risk prediction capability and emergency response efficiency of floating bridge dismantling and anchoring safety in complex water environments. Summary of the Invention

[0005] The technical problem solved by this invention is that existing technologies have difficulty in synchronously collecting and integrating river channel status data in construction waters, and it is difficult to analyze and identify risk patterns of anchorage bearing status.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges, comprising: The data acquisition module is used to perform arch bridge dismantling and cutting construction and obtain river channel status data; The feature construction module is used to process the river state data to obtain the first vector feature and the second vector feature, and then concatenate the vector features to obtain the river feature matrix. The environmental clustering module is used to process the river feature matrix using a clustering algorithm to obtain anchorage bearing capacity assessment data. The safety early warning module is used to render a heat map of the anchorage based on the meta-model, trigger graded alarms, and output construction decision instructions.

[0007] Preferably, the acquisition module includes: The underwater bathymetry phase reference surface was obtained by using an unmanned survey vessel and attitude solver, and spatiotemporal joint calibration. The unmanned surveying vessel was controlled to conduct a spatial grid-based scan along the construction area of ​​the actual river channel, sampling coordinates at each spatial point. At the location, based on the working surface vector of the arch bridge demolition construction, orthogonal dimension detection measurement is performed. The processing logic of the orthogonal dimension detection measurement is as follows: The multibeam sonar is positioned perpendicular to the riverbed interface, transmitting acoustic signals and receiving echoes to collect bottom sediment scattering response data in the water depth direction. The flow profiler is positioned so that its detection direction is parallel to the direction of water flow evolution. It emits acoustic signals and receives echoes to collect flow velocity gradient response data in the horizontal direction of the flow field. The displacement parameters of the floating vessel and the strain parameters of the mooring rod under different working conditions are extracted to construct a multimodal environmental data stream. The spatial coordinates of each spatial sampling coordinate point are mapped to the multimodal environmental data stream to output the river channel state data.

[0008] Preferably, the feature construction module includes a feature extraction unit and a matrix concatenation unit; The feature extraction unit is used to analyze the river state data, extract time-series hydrological evolution features, and construct an ideal hydrodynamic flow field model. The river channel state data of the water-based dismantling and cutting construction environment were analyzed, and the velocity gradient vector, water level fluctuation amplitude, and theoretical scour rate of each time stage were extracted and mapped to first vector features and second vector features: The flow velocity vector and water level data at the hydrodynamic level are extracted as the first vector feature; the riverbed density and initial stress of temporary mooring rods at the geomechanical level are extracted as the second vector feature.

[0009] Preferably, the matrix splicing unit is used to establish the spatiotemporal coupled integral equation of water flow evolution in an ideal flow field model, and the discrete algebraic matrix equation is obtained by discretizing it using a three-dimensional voxel grid and state space interpolation method, and the river channel feature matrix is ​​obtained by vector feature splicing. Under the target construction window, a spatiotemporal coupled integral equation for the evolution of water flow disturbance in an ideal flow field model is established. A three-dimensional voxel grid is used to divide the underwater three-dimensional riverbed digital model, and a state space interpolation method is introduced to discretize the hydrodynamic field. The local velocity field is divided into N linear combinations of local basis functions. The test function method is used to perform a weighted inner product on the integral equation to obtain the discrete algebraic matrix equation. The first vector feature and the second vector feature are aligned by time step within each spatial coordinate block of the three-dimensional voxel grid and then concatenated to obtain the river channel feature matrix. After diagonalizing the channel feature matrix, the theoretical anchoring resistance at the current spatial coordinate point is calculated, and the final output spatial coordinates, channel feature matrix, and theoretical anchoring resistance are stored in the database.

[0010] Preferably, the environment clustering module includes a clustering parsing unit and a field construction unit; The clustering parsing unit is used to reduce the dimensionality of the river channel feature matrix and iterate the center point using a clustering algorithm to obtain the eigenvalues ​​of intermediate variables and the force fluctuation coefficient of the tie rod; Read the river channel feature matrix, perform phase delay correction on the river channel feature matrix, obtain the corrected state feature matrix, use a clustering algorithm to extract the corrected high-dimensional feature center points, calculate the sum and difference variables of hydrological evolution and geological evolution, and obtain the intermediate variable eigenvalues ​​and tie rod stress fluctuation coefficients through the sum and difference variables.

[0011] Preferably, the field construction unit is used to introduce a physical mechanism error compensation strategy, reconstruct the anchorage bearing capacity assessment data through the anchorage force inverse calculation model, extract the absolute value of the actual bearing capacity and calculate the actual anti-sliding anisotropy, and construct the riverbed anchorage physical matrix field. A physical mechanism error compensation strategy is introduced, and the physical relationship between the flow velocity excitation force and the corrected riverbed density is established by utilizing the tie rod force fluctuation coefficient. The inverse calculation model of anchoring force is obtained by using the depth-dominant formula of scour characteristic values; Substituting the hydrodynamic matrix and the geomechanical matrix into the inverse calculation model of anchoring force, the principal bearing capacity and lateral bearing capacity are calculated, and the anchoring bearing capacity assessment data at the current spatial sampling coordinate point is reconstructed. The real part of the main diagonal of the anchoring bearing capacity assessment data is extracted as the absolute value of the actual bearing capacity, and the actual anti-sliding anisotropy at the current spatial sampling coordinate point is calculated. The spatial sampling coordinate points and the corresponding anchoring bearing capacity assessment data are spatially mapped with the actual anti-sliding anisotropy to construct the riverbed anchoring physical matrix field of the entire river channel. Each voxel of the riverbed anchoring physical matrix field includes spatial coordinates, anchoring bearing capacity assessment data, and actual anti-sliding anisotropy.

[0012] Preferably, the safety early warning module includes: A unified three-dimensional spatial coordinate system is constructed by reading the riverbed anchorage physical matrix field and the physical reference database. The actual anchorage bearing capacity assessment data and the theoretical anchorage resistance are matched one-to-one, and the actual anti-sliding anisotropy is matched one-to-one with the reference anisotropy. The bearing characteristic deviation value at each spatial sampling coordinate point in the three-dimensional spatial coordinate system is calculated. The bearing characteristic deviation value includes the absolute bearing value deviation and the anti-sliding degradation degree deviation.

[0013] Preferably, the processing logic for rendering the anchorage bearing heatmap based on the meta-model and triggering a graded alarm is as follows: Establish a mapping rule for judging river channel anomalies, use a meta-model to perform statistical distribution analysis on the deviation data in the global space, and calculate the bearing capacity drop threshold and anti-sliding degradation threshold. River channel condition risks are classified using the bearing capacity drop threshold and the anti-sliding degradation threshold to obtain risk levels. The logic for determining these risk levels is as follows: When the absolute value deviation of bearing capacity is greater than or equal to the bearing capacity drop threshold and the deviation of anti-slip degradation degree is less than the anti-slip degradation threshold, it is determined that there is a level 1 risk in the current coordinate area; When the absolute value deviation of the bearing capacity is less than the bearing capacity drop threshold and the deviation of the anti-slip degradation degree is greater than or equal to the anti-slip degradation threshold, it is determined that there is a level 2 risk in the current coordinate area; When the absolute value deviation of the bearing capacity is greater than or equal to the bearing capacity drop threshold, and the deviation of the anti-slip degradation degree is greater than or equal to the anti-slip degradation threshold, it is determined that there is a level 3 risk in the current coordinate area; When the absolute value deviation of the bearing capacity is less than the bearing capacity drop threshold and the deviation of the anti-slip degradation degree is less than the anti-slip degradation threshold, it is determined that the current area is in a safe and workable zone. The coordinate areas where risks exist are extracted and identified. The local equivalent pull-out force is calculated through quantitative assessment of bearing capacity. The fatigue of the tie rod is quantified in the coordinate areas where risks exist through the flow-induced vibration damage index, and the dynamic damage degree of the tie rod is obtained. The spatial coordinates, risk level, local equivalent pull-out force, and dynamic damage degree of the tie rod at all spatial sampling coordinate points are integrated and marked in a three-dimensional spatial coordinate system. The meta-model is used to render and generate an anchorage bearing heat map and trigger the corresponding graded alarm.

[0014] Preferably, the processing logic for outputting construction decision instructions includes: Read the anchorage bearing heat map, and in the three-dimensional digital twin space, reverse map each spatial sampling coordinate point to the corresponding specific floating vessel compartment and dismantling construction sequence in the multi-dimensional monitoring file of the waterborne dismantling construction environment. Based on the dismantling construction sequence obtained after mapping, implement a differentiated dynamic adjustment strategy for construction technology according to the risk level. The specific processing logic is as follows: When the risk level associated with a construction phase is a safe and workable area, the original cutting and hoisting control commands for the work phase remain unchanged. When the risk level associated with a construction phase is Level 1, a local dynamic ballast water compensation strategy is implemented: Extract the local equivalent pull-out force corresponding to the current water area, calculate the missing gravity compensation of the current floating vessel based on the principle of static equilibrium, increase the water injection volume per unit time of the target ballast tank in the attitude control command at the corresponding location, and write the increased water injection leveling parameter as a positive compensation parameter into the construction decision command. When the risk level associated with the construction sequence is the second level of risk, execute the tension-coordinated tensioning strengthening strategy: Extract the dynamic damage degree of the mooring bar corresponding to the current water area and perform tension field correction processing, which includes prestress correction and buffer displacement correction.

[0015] Preferably, the safety warning module further includes an instruction driving unit: After completing the reverse mapping and parameter modification of all dismantling and cutting construction sequence segments, the updated global scheduling instruction set is subjected to kinematic smoothing filtering to limit the transient and drastic jump rate of water injection valve opening, jack pressure and cutting machine speed instructions between adjacent process segments. The modified construction decision instructions are generated and output, and the modified construction decision instructions are directly sent to the floating automatic ballast terminal with control valve group and the main control machine of the arch bridge water cutting equipment.

[0016] The beneficial effects of this invention are as follows: By using an unmanned survey vessel equipped with multibeam sonar, ADCP, and attitude solver, and simultaneously collecting floating vessel displacement and mooring strain parameters, it is possible to simultaneously acquire flow field distribution, riverbed topography evolution, bottom sediment scattering response, and anchoring system mechanical response during the dismantling and construction process on water. This breaks through the data dimension barriers of traditional methods and provides a complete physical information foundation for safety assessment. Orthogonal dimension detection is performed based on the surface vector of the arch bridge dismantling operation, ensuring that the multibeam sonar detection direction is perpendicular to the riverbed to obtain accurate bottom sediment scattering response, and that the velocity profiler is parallel to the water flow direction to obtain accurate flow field gradient response, thus ensuring the measurement accuracy and physical directionality of data in each dimension. By extracting differentiated features from the hydrodynamic and geomechanical levels, and based on spatiotemporal coupled integral equations and three-dimensional voxel meshes to form a river channel feature matrix, the water flow driving... Dynamics and riverbed scour resistance are integrated into a unified mathematical framework for correlation analysis. Clustering algorithms are used to reduce the dimensionality of high-dimensional feature matrices and iterate the center points. Anchorage bearing capacity assessment data is reconstructed through a physical mechanism error compensation strategy. This approach can automatically identify abnormal patterns in environmental conditions without relying on preset fixed thresholds and is adaptive to novel and complex risks. By inversely calculating the principal and lateral bearing capacities at each spatial coordinate point, the actual anti-sliding anisotropy is calculated, and a full-channel physical matrix field is constructed. This reveals the differences in the bearing capacity of anchors in different directions and provides early warning of directional sliding risks caused by changes in water flow direction or uneven bottom sediment. Using the absolute value deviation of bearing capacity and the deviation of anti-sliding degradation as dual judgment indicators, risks are classified into three levels, and the physical meaning of each type of risk is clearly defined, avoiding false alarms from single-indicator criteria and improving the accuracy of early warning. Attached Figure Description

[0017] Figure 1 This is a basic flowchart of a river condition assessment system for underwater dismantling and cutting construction of an arch bridge, provided as an embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] Example, refer to Figure 1 A river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges is provided, including: The data acquisition module is used to perform arch bridge dismantling and cutting construction and obtain river channel status data; The feature construction module is used to process the river state data to obtain the first vector feature and the second vector feature, and then concatenate the vector features to obtain the river feature matrix. The environmental clustering module is used to process the river feature matrix using a clustering algorithm to obtain anchorage bearing capacity assessment data. The safety early warning module is used to render a heat map of the anchorage based on the meta-model, trigger graded alarms, and output construction decision instructions.

[0020] The acquisition module includes: The underwater bathymetry phase reference surface was obtained by using an unmanned survey vessel and attitude solver, and spatiotemporal joint calibration. The unmanned surveying vessel was controlled to conduct a spatial grid-based scan along the construction area of ​​the actual river channel, sampling coordinates at each spatial point. At the location, based on the working surface vector of the arch bridge demolition construction, orthogonal dimension detection measurement is performed. The processing logic of the orthogonal dimension detection measurement is as follows: The multibeam sonar is positioned perpendicular to the riverbed interface, transmitting acoustic signals and receiving echoes to collect bottom sediment scattering response data in the water depth direction. The flow profiler is positioned so that its detection direction is parallel to the direction of water flow evolution. It emits acoustic signals and receives echoes to collect flow velocity gradient response data in the horizontal direction of the flow field. The displacement parameters of the floating vessel and the strain parameters of the mooring rod under different working conditions are extracted to construct a multimodal environmental data stream. The spatial coordinates of each spatial sampling coordinate point are mapped to the multimodal environmental data stream to output the river channel state data.

[0021] In this embodiment, the data acquisition module utilizes a multibeam sonar, acoustic Doppler current profiler, attitude solver, and displacement and bollard strain sensors mounted on an unmanned survey vessel to acquire river channel state data during the arch bridge dismantling and construction. First, the unmanned survey vessel and the shore-based reference station undergo spatiotemporal joint calibration: using real-time roll, pitch, and heave data provided by the attitude solver, combined with differential satellite positioning, the instantaneous receiving phase of the multibeam sonar is corrected to a unified underwater bathymetry phase reference surface, eliminating the influence of wave undulations on water depth measurement. Subsequently, the unmanned survey vessel was controlled to conduct a spatial scan along the construction water area according to a preset grid path. At each spatial sampling coordinate point (x, y, z), orthogonal dimensional detection and measurement were performed based on the surface vector of the arch bridge demolition operation, i.e., the cutting and propulsion direction and the mooring direction of the floating vessel. The detection beam of the multibeam sonar was rotated to be perpendicular to the riverbed interface (Z-axis), penetrating the water body to emit broadband acoustic signals and receive bottom echoes. The intensity and envelope characteristics of the echoes were extracted, and bottom scattering response data in the water depth direction were collected for inversion of sediment thickness, scour pit morphology, and riverbed bottom density. Simultaneously, the detection beam of the acoustic Doppler current profiler was controlled to be parallel to the direction of water flow evolution (XY plane), emitting acoustic signals and receiving Doppler frequency shift echoes to collect velocity gradient response data in the horizontal direction of the flow field under different water depth stratifications. Furthermore, the displacement parameters of the floating vessel under different working conditions are extracted in real time by differential positioning and inertial measurement units, and the strain parameters of the bollard are obtained by the tension sensor at the bollard end. These data are time-aligned and spatially matched with the corresponding spatial sampling coordinates to construct a multimodal environmental data stream. The final output includes river state data containing velocity gradient, bottom sediment scattering, floating vessel motion, and bollard stress. In the complex underwater demolition of long-span arch bridges, the anchor dragging and instability of floating vessels are often caused by the combined effects of rapid changes in water flow and erosion of the riverbed. This embodiment collects data through orthogonal dimension measurement and performs spatiotemporal alignment on the multimodal data. This embodiment can not only clearly identify the specific reasons for the floating vessel's loss of stability, but also bind the small environmental disturbances (such as the bottom flow velocity change of 0.1 m / s) with the reaction of the construction carrier.

[0022] The feature construction module includes a feature extraction unit and a matrix concatenation unit; The feature extraction unit is used to analyze the river state data, extract time-series hydrological evolution features, and construct an ideal hydrodynamic flow field model. We first analyze the received river state data to extract time-series hydrological evolution features. To establish a benchmark reference system under no construction disturbance conditions, we introduce potential flow theory to construct an ideal hydrodynamic flow field model. Existing flow fields typically rely solely on the standard Laplace equation (…). Theoretical derivation was carried out; however, in this embodiment, in order to prevent the water-blocking constraint effect of the draft of the large floating ship on the water flow section during the demolition of the arch bridge, we modified the boundary conditions and introduced a water-blocking disturbance factor based on the actual water level fluctuation to establish a modified ideal flow field for the construction water area.

[0023] The river channel state data of the water-based dismantling and cutting construction environment were analyzed, and the velocity gradient vector, water level fluctuation amplitude, and theoretical scour rate of each time stage were extracted and mapped to first vector features and second vector features: The flow velocity vector and water level data at the hydrodynamic level are extracted as the first vector feature; the riverbed density and initial stress of temporary mooring rods at the geomechanical level are extracted as the second vector feature.

[0024] Traditional theoretical scour rates typically employ general empirical formulas, considering only the shear force at the bottom of the water flow and the critical initiation shear stress of sediment, neglecting the dynamic disturbance of the riverbed caused by construction equipment. During the dismantling and construction of long-span arch bridges, the minute vibrations of floating vessels can be transmitted to the riverbed interior through temporary moorings and anchorages, leading to micro-liquefaction of sediment or structural loosening in the anchorage area. Therefore, this invention introduces the initial stress of the temporary moorings as a geological attenuation penalty term, constructing an improved spatiotemporally coupled scour rate formula: ; in, This is the improved theoretical scouring rate, in m / s. The erosion coefficient is expressed in units of 1. , The bed shear force is derived from the current flow velocity gradient vector. The initial critical initiation shear stress of the substrate in its natural state. The initial stress of the temporary tie rod was collected. The yield stress of the tie rod material. Geological dynamic load sensitivity coefficient; After completing the above calculations, the theoretical scour rate, etc., are mapped to a first vector feature and a second vector feature. The first vector feature represents the external driving force, and the second vector feature represents the internal resistance to damage. The velocity gradient vector and water level fluctuation data at the hydrodynamic level are extracted and combined to form the first vector feature. The density of the riverbed sediment and the initial stress of the temporary tie rods at the geomechanical level are extracted and combined to form a second vector feature. Traditional scour calculations separate hydrology and construction. The innovation of this embodiment lies in the fact that when dismantling work causes initial stress on the tie rods... When the value is significantly increased, the attenuation term decreases, resulting in a substantial reduction in the equivalent critical shear force against scour of the riverbed sediment, thus leading to a higher theoretical scour rate. This allows the model to effectively simulate the engineering principle that the silt around high-stress anchor points is more easily eroded by flowing water.

[0025] The matrix splicing unit is used to establish the spatiotemporal coupled integral equation of water flow evolution in the ideal flow field model. The discrete algebraic matrix equation is discretized by three-dimensional voxel grid and state space interpolation method, and the river channel feature matrix is ​​obtained by vector feature splicing. Under the target construction window, a spatiotemporal coupled integral equation for the evolution of water flow disturbance in an ideal flow field model is established. A three-dimensional voxel grid is used to divide the underwater three-dimensional riverbed digital model, and a state space interpolation method is introduced to discretize the hydrodynamic field. The local velocity field is divided into N linear combinations of local basis functions. The test function method is used to perform a weighted inner product on the integral equation to obtain the discrete algebraic matrix equation. The first vector feature and the second vector feature are aligned by time step within each spatial coordinate block of the three-dimensional voxel grid and then concatenated to obtain the river channel feature matrix. After diagonalizing the channel feature matrix, the theoretical anchoring resistance at the current spatial coordinate point is calculated. The final output spatial coordinates, channel feature matrix, and theoretical anchoring resistance are then stored in the database. ; ; First, within a pre-defined target construction window, a spatiotemporal coupled integral equation for the evolution of water flow disturbance in an ideal flow field model is established. In this embodiment, the target construction window is preferably set as the critical dismantling period of the next 12 hours. A three-dimensional voxel mesh is used to spatially discretize the underwater three-dimensional riverbed digital model, and a state-space interpolation method is introduced to represent the local velocity field within each voxel. Represented as a linear combination of N local basis functions: Subsequently, a weighted inner product operation is performed on the spatiotemporal coupled integral equation to solve the discrete algebraic matrix equation: ; in Here is the flow field stiffness matrix. Here is the mass inertia matrix. Let be the boundary condition vector that includes riverbed friction and the water-blocking effect of the floating vessel. For vector features, n=1, 2; Given the state feature vectors of each spatially discrete node in a 3D voxel mesh, the first vector feature is extracted within this mesh framework. With the second vector features Because there is a physical lag between the abrupt change in water flow velocity and the response to riverbed density, time step alignment is performed within each three-dimensional voxel grid. After alignment, the channel feature matrix, i.e., the cross-covariance tensor, is constructed. : ; in, Represents the tensor outer product. and This reflects the cross-coupling effect of hydrodynamics on geological resistance. The weighting coefficient is used to characterize the intensity of disturbance during the demolition of the arch bridge.

[0026] Finally, the high-dimensional channel feature matrix is ​​diagonalized, i.e., the eigenvalues ​​and eigenvectors are solved. The largest principal eigenvalue is extracted from its diagonal matrix. This largest principal eigenvalue maps the direction and amplitude of the principal stress with the strongest resistance to damage under the current three-dimensional flow field and substrate conditions. The theoretical anchoring resistance at the current spatial coordinate point is calculated. Finally, the spatial coordinates, the channel feature matrix, and the theoretical anchoring resistance are stored together in the database.

[0027] The environment clustering module includes a clustering parsing unit and a field construction unit; The clustering parsing unit is used to reduce the dimensionality of the river channel feature matrix and iterate the center point using a clustering algorithm to obtain the eigenvalues ​​of intermediate variables and the force fluctuation coefficient of the tie rod; Read the river channel feature matrix, perform phase delay correction on the river channel feature matrix, obtain the corrected state feature matrix, use a clustering algorithm to extract the corrected high-dimensional feature center points, calculate the sum and difference variables of hydrological evolution and geological evolution, and obtain the intermediate variable eigenvalues ​​and tie rod stress fluctuation coefficients through the sum and difference variables.

[0028] In this embodiment, the river channel feature matrix is ​​read and first subjected to phase delay correction: due to the time lag in the responses of each sensor, the peak offset of the cross-correlation function between the velocity vector and the tie rod strain is calculated to perform time alignment correction on each column of the river channel feature matrix, resulting in the corrected state feature matrix. Then, density-based clustering algorithms (such as HDBSCAN and improved K-means++) are used to perform high-dimensional space dimensionality reduction and centroid iteration on the state feature matrix within the continuous time series. The purpose of clustering is to group the chaotic data into several typical operating condition clusters (such as: slow flow stabilization period, rapid flow strong scouring period, anchoring strong disturbance period), and extract the high-dimensional feature centroids of each cluster. And mathematically decompose it into the corresponding normalized hydrological evolution vector subset and the geological stress evolution vector subset; Based on this, this embodiment innovatively calculates the sum and difference variables of hydrological and geological evolution, and further calculates the intermediate variable characteristic value of the core state of the cluster using the above sum and difference variables. The improved formula for the eigenvalue of the intermediate variable, in relation to the force fluctuation coefficient of the tie rod, is as follows: It is used to comprehensively reflect the overall system energy intensity of the current river environment and construction changes; Improved tie rod stress fluctuation coefficient The formula is: ; in, To prevent extremely small smoothing constants with a denominator of zero, this embodiment sets it to... , The elastoplastic nonlinear amplification factor for the tie rod material is used to sensitize and amplify the positive and negative properties of the difference variable using the exponential term; the variable S represents the overall scale, that is, the water flow impact force plus the total resistance of the riverbed and the tie rod. When S is large, it indicates that the current situation is a high-risk confrontation state with high flow velocity and strong force. D represents the degree of imbalance. When the external water flow disturbance continues to increase, and the resistance of the riverbed decreases sharply due to sediment loss, the absolute value of D will increase. The construction period for dismantling arch bridges over water is long, and sensors will generate a large amount of redundant data. This embodiment uses clustering dimensionality reduction, the main purpose of which is feature shrinkage, filtering out normal background noise, and directly extracting the central image of a certain extreme or special working condition, thereby reducing the computing power consumption of the subsequent safety early warning module.

[0029] The field construction unit is used to introduce a physical mechanism error compensation strategy, reconstruct the anchorage bearing capacity assessment data through the anchorage force inverse calculation model, extract the absolute value of the actual bearing capacity and calculate the actual anti-sliding anisotropy, and construct the riverbed anchorage physical matrix field. A physical mechanism error compensation strategy is introduced, and the physical relationship between the flow velocity excitation force and the corrected riverbed density is established by utilizing the tie rod force fluctuation coefficient. The inverse calculation model of anchoring force is obtained by using the depth-dominant formula of scour characteristic values; Substituting the hydrodynamic matrix and the geomechanical matrix into the inverse calculation model of anchoring force, the principal bearing capacity and lateral bearing capacity are calculated, and the anchoring bearing capacity assessment data at the current spatial sampling coordinate point is reconstructed. The real part of the main diagonal of the anchoring bearing capacity assessment data is extracted as the absolute value of the actual bearing capacity, and the actual anti-sliding anisotropy at the current spatial sampling coordinate point is calculated. The spatial sampling coordinate points and the corresponding anchoring bearing capacity assessment data are spatially mapped with the actual anti-sliding anisotropy to construct the riverbed anchoring physical matrix field of the entire river channel. Each voxel of the riverbed anchoring physical matrix field includes spatial coordinates, anchoring bearing capacity assessment data, and actual anti-sliding anisotropy.

[0030] Traditionally, the density of the sediment is usually assumed to be a static constant. However, in actual bridge demolition operations, water flow vibration and tie rod oscillation can cause dynamic liquefaction and weakening of the sediment. This embodiment utilizes the tie rod stress fluctuation coefficient. Establish the relationship between flow velocity excitation force and corrected riverbed density The formula relating them: ; in, The original detection density, The thixotropic damping coefficient for sediment is used in this compensation formula to compensate for severe strut oscillations. When the density is large, even if the water flow excitation force remains unchanged, the equivalent density of the riverbed will decrease nonlinearly. Subsequently, a depth-dominant formula incorporating clustered scour features was used to construct an inverse calculation model for anchorage force. Considering the directionality of bearing capacity in three-dimensional space, the reconstructed anchorage bearing capacity assessment data was defined as a second-order tensor. The improved inverse calculation matrix equation is as follows: : ; in, This is the geomechanical resistance matrix constructed based on the corrected density; This is the hydrodynamic pushover matrix constructed based on the current flow velocity gradient; The equivalent scour depth is calculated from the eigenvalues ​​of intermediate variables; The theoretical effective burial depth for anchoring floating vessels; This is the shape drag coefficient; After substituting the measured hydrodynamic and geomechanical matrices into the above solution model, the anchorage bearing capacity assessment tensor at the current spatial coordinate point is calculated; the real parts of the tensor are extracted from the main diagonal to obtain the principal bearing capacity in the direction of downstream flow evolution. Lateral bearing capacity perpendicular to the direction of water flow ; The principal bearing capacity is selected as the absolute value of the actual bearing capacity at the current coordinate point. Then, the ratio of these two values ​​is calculated to obtain the actual anti-slip anisotropy at the current spatial sampling coordinate point. : ; Finally, the spatial coordinates of each sampling point, the corresponding anchorage bearing capacity assessment data (tensor), the absolute value of the actual bearing capacity and the actual anti-sliding anisotropy are absolutely mapped to spatial coordinates, and the one-dimensional data array is filled into a three-dimensional grid to finally construct a riverbed anchorage physical matrix field covering the entire construction channel; At the moment the arch bridge is dismantled and cut apart, the floating vessel, due to the loss of symmetrical restraint, often experiences extremely large lateral yaw forces or rotational moments. At this time, the main bearing capacity with the downstream current may be very sufficient, but the lateral bearing capacity may be extremely fragile due to the cutting effect of the anchor channel. Design anisotropy is used to quantify the shortcomings of the anchoring system in terms of force in various directions. If the anisotropy is too large, it indicates that the anchoring is very stable with the downstream current, but does not have the ability to resist sideslip.

[0031] The security early warning module includes: A unified three-dimensional spatial coordinate system is constructed by reading the riverbed anchorage physical matrix field and the physical reference database. The actual anchorage bearing capacity assessment data and the theoretical anchorage resistance are matched one-to-one, and the actual anti-sliding anisotropy is matched one-to-one with the reference anisotropy. The bearing characteristic deviation value at each spatial sampling coordinate point in the three-dimensional spatial coordinate system is calculated. The bearing characteristic deviation value includes the absolute bearing value deviation and the anti-sliding degradation degree deviation.

[0032] In this embodiment, the riverbed anchorage physical matrix field constructed by the environmental clustering module and the physical reference database stored by the feature construction module are read and unified into the same three-dimensional spatial coordinate system. In this coordinate system, the actual anchorage bearing capacity assessment data (actual absolute bearing capacity value) of each spatial sampling coordinate point is mapped one-to-one with the theoretical anchorage resistance in the database; simultaneously, the actual anti-sliding anisotropy is mapped one-to-one with the reference anisotropy calculated by the ideal model in the database. Then, the bearing characteristic deviation value at each coordinate point is calculated, including: bearing absolute value deviation (the difference or ratio deviation between the actual absolute bearing capacity value and the theoretical anchorage resistance) and anti-sliding degradation deviation (the degree of deviation between the actual anti-sliding anisotropy and the reference anisotropy). These two deviation values ​​together quantify the current state of the anchorage system relative to the ideal reference.

[0033] The processing logic for rendering the anchorage bearing heat map based on the meta-model and triggering graded alarms is as follows: Establish a mapping rule for judging river channel anomalies, use a meta-model to perform statistical distribution analysis on the deviation data in the global space, and calculate the bearing capacity drop threshold and anti-sliding degradation threshold. River channel condition risks are classified using the bearing capacity drop threshold and the anti-sliding degradation threshold to obtain risk levels. The logic for determining these risk levels is as follows: When the absolute value deviation of bearing capacity is greater than or equal to the bearing capacity drop threshold and the deviation of anti-slip degradation degree is less than the anti-slip degradation threshold, it is determined that there is a level 1 risk in the current coordinate area; When the absolute value deviation of the bearing capacity is less than the bearing capacity drop threshold and the deviation of the anti-slip degradation degree is greater than or equal to the anti-slip degradation threshold, it is determined that there is a level 2 risk in the current coordinate area; When the absolute value deviation of the bearing capacity is greater than or equal to the bearing capacity drop threshold, and the deviation of the anti-slip degradation degree is greater than or equal to the anti-slip degradation threshold, it is determined that there is a level 3 risk in the current coordinate area; When the absolute value deviation of the bearing capacity is less than the bearing capacity drop threshold and the deviation of the anti-slip degradation degree is less than the anti-slip degradation threshold, it is determined that the current area is in a safe and constructible zone; the coordinate areas with risks are extracted and identified, and the local equivalent pull-out force is calculated through quantitative assessment of bearing capacity. The fatigue of the tie rod is quantified for the coordinate areas with risks through the flow-induced vibration damage index to obtain the dynamic damage degree of the tie rod. The spatial coordinates, risk level, local equivalent pull-out force, and dynamic damage degree of the tie rod at all spatial sampling coordinate points are integrated and marked in a three-dimensional spatial coordinate system. The meta-model is used to render and generate an anchorage bearing heat map and trigger the corresponding graded alarm.

[0034] First, a mapping rule for judging river channel anomalies is established. Considering the differences in different water areas and construction seasons, fixed thresholds are prone to false alarms. This embodiment adopts a model based on the gradient boosting tree (GBDT) architecture to perform global probability density statistics on the absolute value deviation of bearing capacity and the deviation of anti-slip degradation in three-dimensional space. By identifying the abrupt inflection point of the density distribution function, the threshold parameters under the current working conditions are adaptively generated. Taking the underwater demolition of a 200-meter main span steel pipe arch bridge as an example, in the initial calibration stage, the automatically generated bearing capacity drop threshold is set to 15% (i.e., the deviation reaches 0.15), and the anti-slip degradation threshold is set to 20% (i.e., the deviation reaches 0.20). Subsequently, based on the aforementioned specific thresholds, each spatial coordinate point is traversed, and cross-judgment is performed to determine the risk level: When the absolute value deviation of the load Furthermore, the deviation in anti-skid degradation is significant. At this time, it is determined that there is a level 1 risk in the current area. This condition indicates that the water flow has caused longitudinal erosion of the bottom sediment, and the total gripping force has decreased, but the direction of force has not been seriously deflected, which is reflected in the erosion of surface sediment and a slight tendency to drift with the current. When the absolute value deviation of the load Furthermore, the deviation in anti-skid degradation is significant. When it is determined that there is a level 2 risk in the current area, the total bearing capacity of the anchorage is within the safe range, but the lateral bearing capacity is severely reduced. This is usually caused by uneven unloading during bridge demolition or vortex-induced resonance of water flow, which manifests as abrupt changes in flow field vibration and lateral imbalance of tie rod tension. When the absolute value deviation of the load And the deviation in anti-slip degradation At that time, it is determined that there is a level 3 risk in the current area, which indicates that the anchoring system not only has insufficient grip but also has serious yaw, with the risk of deep overall slippage failure and uncontrollable drift of the floating vessel. When the absolute value deviation of the load And the deviation in anti-slip degradation When the time comes, determine that the current state is safe; Further quantitative parameter transformation for on-site construction is performed on the extracted risk coordinate area: For areas with Level 1 or Level 3 risk, the initial theoretical pull-out limit of the node is called (in this embodiment, the initial value is set to 800kN). Combined with the current absolute value deviation rate of the bearing capacity, the actual remaining bearing capacity is calculated in reverse. For example, when the bearing deviation reaches 25%, after geometric contact area compensation calculation, the current local equivalent pull-out force is output as 600 kN. For areas with level 2 or 3 risks, the high-frequency alternating amplitude of lateral water flow excitation is extracted. Using the calibrated tie rod theoretical fatigue life (e.g., 10⁶ cycles) as a benchmark, the linear fatigue accumulation law is applied, and an acceleration factor for off-center loading caused by anti-slip degradation is introduced for correction. For example, when the degradation deviation reaches 25%, the current dynamic damage degree of the tie rod is calculated to have increased to 45%, indicating that the tie rod has signs of local yielding or wire breakage.

[0035] Finally, all spatial sampling points are encapsulated, encompassing spatial coordinates, risk level (Level 1 / 2 / 3), local equivalent pull-out force (e.g., 600 kN), and dynamic damage degree of the tie rod (e.g., 45%). In the three-dimensional spatial coordinate system, the meta-model voxel rendering engine is invoked to render the safe zone as green semi-transparent, Level 1 risk as yellow, Level 2 as orange, and Level 3 as red, generating an intuitive three-dimensional anchorage bearing heat map and simultaneously triggering corresponding level audible and visual alarms to the construction control console.

[0036] The processing logic for outputting construction decision instructions includes: Read the anchorage bearing heat map, and in the three-dimensional digital twin space, reverse map each spatial sampling coordinate point to the corresponding specific floating vessel compartment and dismantling construction sequence in the multi-dimensional monitoring file of the waterborne dismantling construction environment. Based on the dismantling construction sequence obtained after mapping, implement a differentiated dynamic adjustment strategy for construction technology according to the risk level. The specific processing logic is as follows: When the risk level associated with a construction phase is a safe and workable area, the original cutting and hoisting control commands for the work phase remain unchanged. When the risk level associated with a construction phase is Level 1, a local dynamic ballast water compensation strategy is implemented: Extract the local equivalent pull-out force corresponding to the current water area, calculate the missing gravity compensation of the current floating vessel based on the principle of static equilibrium, increase the water injection volume per unit time of the target ballast tank in the attitude control command at the corresponding location, and write the increased water injection leveling parameter as a positive compensation parameter into the construction decision command. When the risk level associated with the construction sequence is the second level of risk, execute the tension-coordinated tensioning strengthening strategy: Extract the dynamic damage degree of the mooring bar corresponding to the current water area and perform tension field correction processing, which includes prestress correction and buffer displacement correction.

[0037] In this embodiment, when outputting construction decision commands, the anchorage bearing heat map is first read. In a pre-established three-dimensional digital twin space, each spatial sampling coordinate point is reverse-mapped to a specific floating vessel compartment (such as the ballast water tank number) and a dismantling construction sequence (such as the i-th segment cutting operation of the arch rib). Based on the risk level associated with the dismantling construction sequence obtained after mapping, a differentiated dynamic adjustment strategy for the construction process is executed. When the risk level is safe, the original cutting and hoisting control commands remain unchanged. When it is the first level of risk, a local ballast water dynamic compensation strategy is executed: the local equivalent pull-out force corresponding to the water area is extracted, the gravity compensation amount required for the current floating vessel to restore anchorage force is calculated based on the principle of static equilibrium, the unit time water injection volume of the target ballast water tank is added to the attitude control command of the corresponding compartment, and the increased water injection leveling parameter is written into the construction decision command as a positive compensation parameter. When the risk level is Level 2, a tension-coordinated tensioning strengthening strategy is implemented: the dynamic damage degree of the mooring rod corresponding to the water area is extracted, and the tension field is corrected, including increasing the prestress setting value of the mooring rod according to the damage degree, and adding a buffer displacement correction amount to the original control command to suppress the vibration amplitude of the mooring rod.

[0038] The security early warning module also includes an instruction driving unit: After completing the reverse mapping and parameter modification of all dismantling and cutting construction sequence segments, the updated global scheduling instruction set is subjected to kinematic smoothing filtering to limit the transient and drastic jump rate of water injection valve opening, jack pressure and cutting machine speed instructions between adjacent process segments. The modified construction decision instructions are generated and output, and the modified construction decision instructions are directly sent to the floating automatic ballast terminal with control valve group and the main control machine of the arch bridge water cutting equipment.

[0039] This embodiment obtains an updated global scheduling instruction set after completing the reverse mapping and parameter modification of all dismantling and cutting construction sequence segments. To avoid impacts on the floating vessel and cutting equipment caused by parameter mutations, the global instruction set undergoes kinematic smoothing filtering: limiting the transient and drastic jump rate of water injection valve opening, jack pressure, and cutting machine speed commands between adjacent process segments, and using S-shaped acceleration / deceleration curves or low-pass filters to smooth the transition of the instruction sequence. Finally, the corrected construction decision commands are generated and output, and directly sent to the floating vessel automatic ballast terminal with control valve groups and the main control unit of the arch bridge waterborne cutting equipment via industrial Ethernet to achieve closed-loop control.

[0040] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0041] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0042] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

Claims

1. A river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges, characterized in that, include: The data acquisition module is used to perform arch bridge dismantling and cutting construction and obtain river channel status data; The feature construction module is used to process the river state data to obtain the first vector feature and the second vector feature, and then concatenate the vector features to obtain the river feature matrix. The environmental clustering module is used to process the river feature matrix using a clustering algorithm to obtain anchorage bearing capacity assessment data. The safety early warning module is used to draw and render a heat map of the anchorage based on the meta-model, trigger graded alarms, and output construction decision instructions.

2. The river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges as described in claim 1, characterized in that, The acquisition module includes: The underwater bathymetry phase reference surface was obtained by using an unmanned survey vessel and attitude solver, and spatiotemporal joint calibration. The unmanned surveying vessel was controlled to conduct a spatial grid-based scan along the construction area of ​​the actual river channel, sampling coordinates at each spatial point. At the location, based on the working surface vector of the arch bridge demolition construction, orthogonal dimension detection measurement is performed. The processing logic of the orthogonal dimension detection measurement is as follows: The multibeam sonar is positioned perpendicular to the riverbed interface, transmitting acoustic signals and receiving echoes to collect bottom sediment scattering response data in the water depth direction. The flow profiler is positioned so that its detection direction is parallel to the direction of water flow evolution. It emits acoustic signals and receives echoes to collect flow velocity gradient response data in the horizontal direction of the flow field. The displacement parameters of the floating vessel and the strain parameters of the mooring rod under different working conditions are extracted to construct a multimodal environmental data stream. The spatial coordinates of each spatial sampling coordinate point are mapped to the multimodal environmental data stream to output the river channel state data.

3. The river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges as described in claim 2, characterized in that, The feature construction module includes a feature extraction unit and a matrix concatenation unit; The feature extraction unit is used to analyze the river state data, extract time-series hydrological evolution features, and construct an ideal hydrodynamic flow field model. The river channel state data of the water-based dismantling and cutting construction environment were analyzed, and the velocity gradient vector, water level fluctuation amplitude, and theoretical scour rate of each time stage were extracted and mapped to first vector features and second vector features: The flow velocity vector and water level data at the hydrodynamic level are extracted as the first vector feature; the riverbed density and initial stress of temporary mooring rods at the geomechanical level are extracted as the second vector feature.

4. The river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges as described in claim 3, characterized in that, The matrix splicing unit is used to establish the spatiotemporal coupled integral equation of water flow evolution in the ideal flow field model. The discrete algebraic matrix equation is discretized by three-dimensional voxel grid and state space interpolation method, and the river channel feature matrix is ​​obtained by vector feature splicing. Under the target construction window, a spatiotemporal coupled integral equation for the evolution of water flow disturbance in an ideal flow field model is established. A three-dimensional voxel grid is used to divide the underwater three-dimensional riverbed digital model, and a state space interpolation method is introduced to discretize the hydrodynamic field. The local velocity field is divided into N linear combinations of local basis functions. The test function method is used to perform a weighted inner product on the integral equation to obtain the discrete algebraic matrix equation. The first vector feature and the second vector feature are aligned by time step within each spatial coordinate block of the three-dimensional voxel grid and then concatenated to obtain the river channel feature matrix. After diagonalizing the channel feature matrix, the theoretical anchoring resistance at the current spatial coordinate point is calculated, and the final output spatial coordinates, channel feature matrix, and theoretical anchoring resistance are stored in the database.

5. The river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges as described in claim 4, characterized in that, The environment clustering module includes a clustering parsing unit and a field construction unit; The clustering parsing unit is used to reduce the dimensionality of the river channel feature matrix and iterate the center point using a clustering algorithm to obtain the eigenvalues ​​of intermediate variables and the force fluctuation coefficient of the tie rod; Phase delay correction is performed on the river channel feature matrix to obtain the corrected state feature matrix. Clustering algorithm is used to extract the corrected high-dimensional feature center points. The sum and difference variables of hydrological evolution and geological evolution are calculated. The eigenvalues ​​of intermediate variables and the force fluctuation coefficient of tie rods are obtained by calculating the sum and difference variables.

6. The river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges as described in claim 5, characterized in that, The field construction unit is used to introduce a physical error compensation strategy, reconstruct the anchorage bearing capacity assessment data through the anchorage force inverse calculation model, extract the absolute value of the actual bearing capacity and calculate the actual anti-sliding anisotropy, and construct the riverbed anchorage physical matrix field. A physical mechanism error compensation strategy is introduced, and the physical relationship between the flow velocity excitation force and the corrected riverbed density is established by utilizing the tie rod force fluctuation coefficient. The inverse calculation model of anchoring force is obtained by using the depth-dominant formula of scour characteristic values; Substituting the hydrodynamic matrix and the geomechanical matrix into the inverse calculation model of anchoring force, the principal bearing capacity and lateral bearing capacity are calculated, and the anchoring bearing capacity assessment data at the current spatial sampling coordinate point is reconstructed. The real part of the main diagonal of the anchoring bearing capacity assessment data is extracted as the absolute value of the actual bearing capacity, and the actual anti-sliding anisotropy at the current spatial sampling coordinate point is calculated. The spatial sampling coordinate points and the corresponding anchoring bearing capacity assessment data are spatially mapped with the actual anti-sliding anisotropy to construct the riverbed anchoring physical matrix field of the entire river channel. Each voxel of the riverbed anchoring physical matrix field includes spatial coordinates, anchoring bearing capacity assessment data, and actual anti-sliding anisotropy.

7. The river channel condition assessment system for the underwater dismantling and cutting construction of an arch bridge as described in claim 6, characterized in that, The security early warning module includes: A unified three-dimensional spatial coordinate system is constructed by reading the riverbed anchorage physical matrix field and the physical reference database. The actual anchorage bearing capacity assessment data and the theoretical anchorage resistance are matched one-to-one, and the actual anti-sliding anisotropy is matched one-to-one with the reference anisotropy. The bearing characteristic deviation value at each spatial sampling coordinate point in the three-dimensional spatial coordinate system is calculated. The bearing characteristic deviation value includes the absolute bearing value deviation and the anti-sliding degradation degree deviation.

8. The river channel condition assessment system for the underwater dismantling and cutting construction of arch bridges as described in claim 7, characterized in that, The processing logic for rendering a heat map of the anchorage based on the meta-model and triggering a graded alarm is as follows: Establish a mapping rule for judging river channel anomalies, use a meta-model to perform statistical distribution analysis on the deviation data in the global space, and calculate the bearing capacity drop threshold and anti-sliding degradation threshold. River channel condition risks are classified using the bearing capacity drop threshold and the anti-sliding degradation threshold to obtain risk levels. The logic for determining these risk levels is as follows: When the absolute value deviation of bearing capacity is greater than or equal to the bearing capacity drop threshold and the deviation of anti-slip degradation degree is less than the anti-slip degradation threshold, it is determined that there is a level 1 risk in the current coordinate area; When the absolute value deviation of the bearing capacity is less than the bearing capacity drop threshold and the deviation of the anti-slip degradation degree is greater than or equal to the anti-slip degradation threshold, it is determined that there is a level 2 risk in the current coordinate area; When the absolute value deviation of the bearing capacity is greater than or equal to the bearing capacity drop threshold, and the deviation of the anti-slip degradation degree is greater than or equal to the anti-slip degradation threshold, it is determined that there is a level 3 risk in the current coordinate area; When the absolute value deviation of bearing capacity is less than the bearing capacity drop threshold and the deviation of anti-slip degradation degree is less than the anti-slip degradation threshold, it is determined that the current area is in a safe and workable zone. The first level of risk is used to indicate that there is surface mud and sand erosion and slight anchor drift tendency in the current anchoring area. The second level of risk indicates that there are sudden changes in the flow field and imbalances in the tie rod tension in the current water area; The third level of risk indicates that there is a risk of deep slip failure and large-scale drift of the floating vessel in the current coordinate area. The coordinate areas where risks exist are extracted and identified. The local equivalent pull-out force is calculated through quantitative assessment of bearing capacity. The fatigue of the tie rod is quantified in the coordinate areas where risks exist through the flow-induced vibration damage index, and the dynamic damage degree of the tie rod is obtained. The spatial coordinates, risk level, local equivalent pull-out force, and dynamic damage degree of the tie rod at all spatial sampling coordinate points are integrated and marked in a three-dimensional spatial coordinate system. The meta-model is used to render and generate an anchorage bearing heat map and trigger the corresponding graded alarm.

9. A river channel condition assessment system for the underwater dismantling and cutting construction of an arch bridge as described in claim 8, characterized in that, The processing logic for outputting construction decision instructions includes: Read the anchorage bearing heat map, and in the three-dimensional digital twin space, reverse map each spatial sampling coordinate point to the corresponding specific floating vessel compartment and dismantling construction sequence in the multi-dimensional monitoring file of the waterborne dismantling construction environment. Based on the dismantling construction sequence obtained after mapping, implement a differentiated dynamic adjustment strategy for construction technology according to the risk level. The specific processing logic is as follows: When the risk level associated with a construction phase is a safe and workable area, the original cutting and hoisting control commands for the work phase remain unchanged. When the risk level associated with a construction phase is Level 1, a local dynamic ballast water compensation strategy is implemented: Extract the local equivalent pull-out force corresponding to the current water area, calculate the missing gravity compensation of the current floating vessel based on the principle of static equilibrium, increase the water injection volume per unit time of the target ballast tank in the attitude control command at the corresponding location, and write the increased water injection leveling parameter as a positive compensation parameter into the construction decision command. When the risk level associated with the construction sequence is the second level of risk, execute the tension-coordinated tensioning strengthening strategy: Extract the dynamic damage degree of the mooring bar corresponding to the current water area and perform tension field correction processing, which includes prestress correction and buffer displacement correction.

10. The river channel condition assessment system for the underwater dismantling and cutting construction of an arch bridge as described in claim 9, characterized in that, The security early warning module also includes an instruction driving unit: After completing the reverse mapping and parameter modification of all dismantling and cutting construction sequence segments, the updated global scheduling instruction set is subjected to kinematic smoothing filtering to limit the transient and drastic jump rate of water injection valve opening, jack pressure and cutting machine speed instructions between adjacent process segments. The modified construction decision instructions are generated and output, and the modified construction decision instructions are directly sent to the floating automatic ballast terminal with control valve group and the main control machine of the arch bridge water cutting equipment.

Images