A dam safety monitoring system and method based on digital twin

Abstract

The present invention discloses a dam safety monitoring system and method based on digital twin, belonging to the technical field of data processing. It includes generating a multi-dimensional damage feature matrix by collecting surface acoustic emission signals and internal infrared thermal image data of the dam in real time through multi-source sensors; constructing a disaster node topological network by combining density peak clustering and dynamic weights; accurately predicting the damage chain reaction path based on gravitational field strength calculation and improved Monte Carlo simulation; generating a dynamic risk heat map through space-time folding mapping and bidirectional diffraction equation; extracting hidden danger indicators by multi-axis domain feature distillation, dynamically comparing the safety envelope function, and online generating flood discharge scheduling and reinforcement plans. The present invention adopts cross-modal fusion, gravitational field dynamic topological network and space-time folding mapping technology, which can realize the deep integration of accurate simulation of damage energy conduction path, carbonization-deformation bidirectional coupling prediction and intelligent disposal decision-making of risk hidden dangers, and improve the accuracy and initiative of dam safety early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a dam safety monitoring system and method based on digital twin. Background Art

[0002] Traditional dam safety monitoring technologies mostly rely on the discretized measurement of a single physical quantity, such as the monitoring of local parameters like displacement sensors and piezometers. There are problems such as difficulties in fusing multi-source heterogeneous data and insufficient revelation of potential damage evolution mechanisms. Existing methods are difficult to achieve cross-scale dynamic coupling analysis of dam structure damage, resulting in limited prediction accuracy for the hidden danger chain reaction path. Although digital twin technology has been applied to dam monitoring, existing digital models mostly focus on geometric information mapping and data visualization, and fail to deeply integrate algorithm innovations such as cross-modal convolution kernels, dynamic topology of the gravitational field, and spatio-temporal folding mapping. As a result, the twin body cannot truly reflect the catastrophic process under the coupling action of multiple physical fields. Therefore, there is an urgent need to construct an integrated monitoring system that can connect the evolution of microscopic damage mechanisms and the dynamic assessment of macroscopic risks to improve the accuracy and initiative of dam safety early warning. Summary of the Invention

[0003] To solve the above problems, the present invention provides a dam safety monitoring system and method based on digital twin. By using cross-modal fusion, dynamic topology network of the gravitational field, and spatio-temporal folding mapping technologies, it can achieve the deep integration of accurate simulation of damage energy conduction paths, bidirectional coupling prediction of carbonization-deformation, and intelligent disposal decision-making for risk hazards, improving the accuracy and initiative of dam safety early warning.

[0004] The above object can be achieved through the following solutions:

[0005] A dam safety monitoring method based on digital twin, comprising: obtaining the acoustic emission signal distribution map on the dam surface and the internal infrared thermal image matrix, and generating a multi-dimensional damage feature matrix containing the strain energy density field through cross-modal convolution kernel fusion calculation; establishing a disaster node-based topological network based on the spatial coordinates of the multi-dimensional damage feature matrix; inputting the node weights in the topological network into the gravitational field strength calculation model, simulating the energy conduction path, and outputting a set of chain reaction probability values between adjacent nodes in the path; according to the set of chain reaction probability values, performing spatio-temporal folding mapping operation on the multi-dimensional damage feature matrix, establishing a bidirectional diffraction equation between the daily change rate of the microscopic carbonation depth and the cumulative macroscopic displacement, and generating a carbonation-deformation coupling prediction curve; based on the carbonation-deformation coupling prediction curve, calculating the damage energy exchange rate at each monitoring point, performing non-linear Gram matrix multiplication on the energy exchange rate and the chain reaction probability value at the corresponding position, and generating a standardized risk coupling coefficient heat map; performing multi-axis domain feature distillation on the risk coupling coefficient heat map, extracting hidden danger quantification indicators including the mutation gradient of the gravitational field strength, the seepage phase spectrum entropy, and the approaching energy convergence degree, and dynamically comparing them with the safety envelope function in the knowledge base; when any hidden danger quantification indicator breaks through the preset envelope boundary, generating a disposal plan including a microscopic repair coordinate set, a flood discharge path rescheduling sequence, and a reinforcement strength vector map.

[0006] Optionally, the cross-modal convolution kernel fusion calculation includes: performing wavelet packet decomposition on the acoustic emission signal distribution map to generate a multi-band energy distribution tensor; performing edge enhancement on the internal infrared thermal image matrix based on the morphological gradient operator, and outputting a spatial gradient matrix of the heat conduction abnormal area; inputting the multi-band energy distribution tensor and the spatial gradient matrix into a two-channel residual network for cross-modal feature fusion, and optimizing the fusion weight through the strain energy density field loss function to generate the multi-dimensional damage feature matrix.

[0007] Optionally, the establishment of the disaster node-based topological network includes: based on the strain energy density field distribution of the multi-dimensional damage feature matrix, using the density peak clustering algorithm to screen the potential disaster node set; performing Delaunay hierarchical dissection on the node set, constructing a three-dimensional space grid model and dividing it into multiple grid cells; calculating the energy gradient change rate between adjacent grid cells to generate a dynamic weight matrix; using the dynamic weight matrix to output a weighted disaster node-based topological network.

[0008] Optionally, the simulation of the energy conduction path includes: inputting the dynamic weight matrix of the topological network into the gravitational field model to calculate the potential energy field matrix between nodes; using the potential energy field matrix and the node set, and adopting an improved Markov chain Monte Carlo algorithm to simulate the energy conduction path, outputting a set of chain reaction probability values between adjacent nodes, and storing them in the dynamic risk probability database.

[0009] Optionally, the spatio-temporal folding mapping operation includes: framing the multi-dimensional damage feature matrix according to a time series, combining the spatio-temporal distribution of the set of chain reaction probability values to construct a four-dimensional spatio-temporal hypercube; performing a discrete cosine transform on the carbonation depth axis of the four-dimensional spatio-temporal hypercube to generate a frequency-domain projection subspace; performing a second-order difference calculation on the macroscopic displacement cumulative amount axis of the four-dimensional spatio-temporal hypercube to extract frequency-domain and time-domain features; solving the coupling coefficient tensor of the bidirectional diffraction equation to generate a relationship curve between the carbonation depth change rate and the macroscopic displacement, and obtaining a carbonation-deformation coupling prediction curve.

[0010] Optionally, the calculation of the damage energy exchange rate includes: extracting the instantaneous slope sequence of each monitoring point along the carbonation-deformation coupling prediction curve; aligning the instantaneous slope sequence with the set of chain reaction probability values according to the topological network node coding; fusing the environmental temperature and humidity time series data based on the Lagrangian interpolation algorithm for the coded and aligned data to generate a dynamic exchange rate correction factor; eliminating noise interference through the principal component analysis method and outputting the damage energy exchange rate.

[0011] Optionally, the generation of the standardized risk coupling coefficient heat map includes: mapping the damage energy exchange rate and the chain reaction probability value to the grid vertices of the topological network; constructing a three-dimensional covariance Gram matrix and strengthening the non-linear coupling effect through the hyperbolic tangent activation function; using the t-SNE algorithm in manifold learning to reduce the dimension to a two-dimensional plane; dynamically superimposing the risk coefficient distribution according to the time label of the spatio-temporal hypercube and outputting the standardized heat map.

[0012] Optionally, the multi-axis domain feature distillation includes: calculating the directional differential operator along the heat gradient axis of the standardized risk coupling coefficient heat map to extract the geometric feature vector of the gravitational field strength mutation region; performing a fast Fourier transform on the heat frames of continuous time periods to calculate the power spectrum entropy as the percolation phase spectrum entropy; fitting the Lyapunov exponent by combining the energy convergence trajectory of the spatio-temporal hypercube to generate the approaching energy convergence degree; outputting a set of hidden danger quantification indexes including the mutation gradient, spectrum entropy and convergence degree.

[0013] Optionally, the dynamic comparison includes: inputting the set of hidden danger quantification indexes into the fuzzy Petri net inference engine in the knowledge base; matching the critical interval of the gravitational field strength elastic modulus and the percolation phase safety domain in the safety envelope function; when any index breaks through the envelope boundary, triggering the Monte Carlo tree search protocol of the three-dimensional decision engine; outputting the reinforcement strength vector diagram and the flood discharge scheduling sequence based on the optimal convergence solution of the energy conduction path and the coupling prediction curve.

[0014] Based on the same inventive concept, the present invention also provides a dam safety monitoring system based on digital twin. The system includes: a feature acquisition module for obtaining the acoustic emission signal distribution map on the dam surface and the internal infrared thermal image matrix, and generating a multi-dimensional damage feature matrix including the strain energy density field through cross-modal convolution kernel fusion calculation; a disaster network construction module for establishing a disaster node-based topological network based on the spatial coordinates of the multi-dimensional damage feature matrix; a reaction probability calculation module for inputting the node weights in the topological network into the gravitational field strength calculation model, simulating the energy conduction path, and outputting a set of chain reaction probability values between adjacent nodes in the path; a carbonation-deformation coupling prediction curve generation module for performing spatio-temporal folding mapping operation on the multi-dimensional damage feature matrix according to the set of chain reaction probability values, establishing a bidirectional diffraction equation between the daily change rate sequence of the microscopic carbonation depth and the cumulative macroscopic displacement, and generating a carbonation-deformation coupling prediction curve; a risk coupling coefficient heat map generation module for calculating the damage energy exchange rate at each monitoring point based on the carbonation-deformation coupling prediction curve, performing non-linear Gram matrix multiplication on the energy exchange rate and the chain reaction probability value at the corresponding position, and generating a standardized risk coupling coefficient heat map; a hidden danger quantification index extraction module for performing multi-axis domain feature distillation on the risk coupling coefficient heat map, extracting hidden danger quantification indexes including the mutation gradient of the gravitational field strength, the seepage phase spectrum entropy, and the approaching energy convergence degree, and dynamically comparing them with the safety envelope function in the knowledge base; a solution generation module for generating a disposal solution including a microscopic repair coordinate set, a flood discharge path rescheduling sequence, and a reinforcement strength vector map when any hidden danger quantification index breaks through the preset envelope boundary.

[0015] Compared with the prior art, the present invention has the following advantages:

[0016] 1. The present invention obtains the acoustic emission signal distribution map on the dam surface and the internal infrared thermal image matrix, and generates a multi-dimensional damage feature matrix including the strain energy density field through cross-modal convolution kernel fusion calculation; this method effectively fuses multi-source heterogeneous data, overcomes the limitations of traditional single physical quantity monitoring, and provides more comprehensive and accurate damage information;

[0017] 2. Based on the spatial coordinates of the multi-dimensional damage feature matrix, the present invention establishes a disaster node-based topological network and simulates the energy conduction path through the gravitational field strength calculation model; this method can accurately simulate the propagation process of damage inside the dam and provides a scientific basis for predicting potential disasters;

[0018] 3. The present invention establishes a bidirectional diffraction equation between the daily change rate sequence of the microscopic carbonation depth and the cumulative macroscopic displacement through spatio-temporal folding mapping operation, and generates a carbonation-deformation coupling prediction curve; this bidirectional coupling analysis can more accurately reveal the dam damage evolution mechanism and improve the prediction accuracy;

[0019] 4. When any hidden danger quantification index breaks through the preset envelope boundary, the present invention can automatically generate a disposal plan including a microscopic repair coordinate set, a flood discharge path rescheduling sequence, and a reinforcement strength vector diagram; this intelligent disposal decision-making mechanism can take effective measures in a timely manner when there are safety hazards in the dam to avoid disasters.

[0020] Other features and advantages of the present invention will be described in the following specification, and, in part, will become apparent from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures pointed out in the specification, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0022] Figure 1 is a schematic flow chart of a digital twin-based dam safety monitoring method according to an embodiment of the present invention.

[0023] Figure 2 is a schematic diagram of a carbonization-deformation coupling prediction curve according to an embodiment of the present invention.

[0024] Figure 3 is a schematic structural diagram of a digital twin-based dam safety monitoring system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0026] Referring to Figure 1 , an embodiment of the present invention proposes a digital twin-based dam safety monitoring method, which adopts cross-modal fusion, gravitational field dynamic topology network, and spatio-temporal folding mapping technologies, and can realize the in-depth integration of accurate simulation of damage energy conduction paths, carbonization-deformation bidirectional coupling prediction, and intelligent disposal decision-making of risk hazards, improving the accuracy and initiative of dam safety early warning.

[0027] The method of this embodiment specifically includes:

[0028] Obtain the acoustic emission signal distribution map on the surface of the dam and the internal infrared thermal image matrix, and generate a multi-dimensional damage feature matrix containing the strain energy density field through cross-modal convolution kernel fusion calculation;

[0029] Based on the spatial coordinates of the multi-dimensional damage feature matrix, establish a disaster node-based topological network;

[0030] Input the node weights in the topological network into the gravitational field strength calculation model, simulate the energy conduction path, and output a set of chain reaction probability values between adjacent nodes in the path;

[0031] According to the set of chain reaction probability values, perform spatio-temporal folding mapping operation on the multi-dimensional damage feature matrix, establish a bidirectional diffraction equation between the daily change rate sequence of the microscopic carbonation depth and the cumulative macro displacement, and generate a carbonation-deformation coupling prediction curve;

[0032] Based on the carbonation-deformation coupling prediction curve, calculate the damage energy exchange rate at each monitoring point, and perform non-linear Gram matrix multiplication on the energy exchange rate and the chain reaction probability value at the corresponding position to generate a standardized risk coupling coefficient heat map;

[0033] Perform multi-axis domain feature distillation on the risk coupling coefficient heat map, extract hidden danger quantification indicators including the mutation gradient of the gravitational field strength, the spectral entropy of the seepage phase, and the approaching energy convergence degree, and perform dynamic comparison with the safety envelope function in the knowledge base;

[0034] When any hidden danger quantification indicator breaks through the preset envelope boundary, generate a disposal plan including a microscopic repair coordinate set, a flood discharge path rescheduling sequence, and a reinforcement strength vector map.

[0035] Specifically, this method establishes a digital twin through cross-modal fusion, dynamic topological modeling, and multi-dimensional spatio-temporal analysis to achieve the connection between the damage evolution mechanism and the macro risk. The principle is to map multi-source data such as acoustic emission and infrared thermal imaging into a unified damage feature matrix, construct a weighted topological network to simulate the energy conduction path, use spatio-temporal mapping and bidirectional diffraction equation to couple microscopic carbonation and macro deformation parameters, and distill key hidden danger indicators through the risk heat map. The technical effects are reflected in improving the simulation accuracy of the damage propagation path, enhancing the spatio-temporal correlation of hidden danger prediction, realizing the bidirectional coupling analysis of carbonation and deformation, and being able to actively generate targeted disposal plans to improve the early warning ability and response efficiency of dam safety monitoring.

[0036] Optionally, the cross-modal convolution kernel fusion calculation includes:

[0037] Perform wavelet packet decomposition on the acoustic emission signal distribution map to generate a multi-band energy distribution tensor;

[0038] Specifically, wavelet packet decomposition is a signal processing method. By performing a more refined frequency band division on the acoustic emission signal distribution map, the original signal is decomposed into sub-signals covering different frequency bands, generating a multi-band energy distribution tensor. This tensor is a three-dimensional data structure used to record the distribution intensity of energy in each frequency band at different spatial positions. Multi-band energy analysis can capture low-frequency fracture waves and high-frequency micro-crack signals generated during the material damage process, providing a basis for the feature recognition of different damage modes.

[0039] Perform edge enhancement on the internal infrared thermal image matrix based on the morphological gradient operator to output the spatial gradient matrix of the heat conduction anomaly region;

[0040] Specifically, edge enhancement based on the morphological gradient operator is an image processing technique. The morphological gradient operator strengthens the temperature gradient change characteristics of the heat conduction anomaly region by calculating the difference between dilation and erosion in the local region of the infrared thermal image matrix, and outputs the spatial gradient matrix. The numerical value of this matrix represents the severity of temperature change at each pixel point. High gradient values correspond to potential regions of heat conduction obstruction or material defects, and can effectively locate internal damage below the surface layer.

[0041] Input the multi-band energy distribution tensor and the spatial gradient matrix into a dual-channel residual network for cross-modal feature fusion, and optimize the fusion weights through the strain energy density field loss function to generate the multi-dimensional damage feature matrix.

[0042] Specifically, the dual-channel residual network is a deep learning architecture with dual-input branches that can process the acoustic emission multi-band energy distribution tensor and the infrared thermal image spatial gradient matrix in parallel. During the feature fusion stage, the detailed information of the original modality is retained through residual connections. The strain energy density field loss function is the optimization objective function during network training. By calculating the energy distribution difference between the predicted strain energy density field and the actual physical model, the fusion weights of cross-modal features are dynamically adjusted. The finally generated multi-dimensional damage feature matrix integrates the frequency-domain energy features of the acoustic emission signal and the spatial temperature gradient features of the infrared thermal image.

[0043] Among them, the input layer of the dual-channel residual network is the acoustic emission tensor (64×64×8) and the infrared gradient matrix (256×256), which are downsampled to the same dimension by a convolution kernel of 3×3 with a stride of 2; the residual block contains 4 residual units per channel. During the cross-modal fusion stage, the number of channels is aligned through 1×1 convolution kernels, and the fusion weights are optimized by backpropagation of the strain energy density loss function; the training parameters are a Dropout rate of 0.3, an initial learning rate of 0.001 for the Adam optimizer, a batch size of 32, and the training set is a cross-modal image database containing 200 cases of dam damage.

[0044] Among them, the strain energy density field loss function is defined as:

[0045] ,

[0046] In the formula, is the predicted strain energy density value corresponding to the th grid cell, is the measured value of the actual strain energy density of the th grid cell, is the total variation regularization coefficient, is the total variation regularization term, is the number of grid cells.

[0047] Specifically, the damage characterization ability is improved through the refined processing of cross-modal data and deep feature fusion. The principle lies in using wavelet packet decomposition to analyze the frequency domain characteristics of acoustic emission signals, combining morphological gradient operators to extract the spatial anomaly features of thermal image data, and fusing heterogeneous modal information through a dual-channel network. The technical effect is reflected in enhancing the capture accuracy of weak signals in the early stage of material damage, suppressing the environmental noise interference of single-sensor data, ensuring that the fused multi-dimensional damage feature matrix contains both frequency domain energy distribution and spatial thermodynamics anomaly information, and providing a more accurate input basis for subsequent topological network construction and risk prediction.

[0048] Optionally, the establishment of the disaster node-based topological network includes:

[0049] Based on the strain energy density field distribution of the multi-dimensional damage feature matrix, the density peak clustering algorithm is used to screen the potential disaster node set;

[0050] Specifically, the density peak clustering algorithm is an unsupervised learning method based on the local density and relative distance of data points. By calculating the local density of each region in the strain energy density field and the minimum distance to higher density points, the potential disaster node set is screened out, that is, the damage core region with high strain energy density and relatively isolated. This algorithm can automatically identify high-energy aggregation points and exclude low-density noise interference, ensuring that the node set covers the real potential damage positions.

[0051] Among them, in the density peak clustering algorithm, the neighborhood radius is set according to the geometric size of the dam, and the neighborhood radius is 0.1 times the maximum span of the dam.

[0052] Perform Delaunay hierarchical triangulation on the node set, construct a three-dimensional space grid model and divide it into multiple grid cells;

[0053] Specifically, Delaunay hierarchical dissection is a three-dimensional space meshing method that generates a non-overlapping tetrahedral element network for the points concentrated with disaster nodes according to the triangulation criterion, and constructs a three-dimensional space grid model with optimized structure. Each grid unit ensures that any point in space is not located inside the circumscribed sphere of other tetrahedrons through the Delaunay condition. The formed grid topology has geometric stability and can accurately describe the spatial distribution characteristics of damaged nodes.

[0054] Calculate the energy gradient change rate between adjacent grid units to generate a dynamic weight matrix;

[0055] Specifically, the energy gradient change rate is an index for calculating the change rate of strain energy density between adjacent grid units. By comparing the difference in average strain energy density within adjacent tetrahedral elements and normalizing by the grid side length, a dynamic weight matrix is generated. This matrix quantifies the energy transfer intensity between nodes, and the weight value reflects the driving potential for damage to spread from high-energy units to low-energy units, endowing the topological network with dynamic evolution characteristics.

[0056] Use the dynamic weight matrix to output a weighted disaster node-based topological network.

[0057] Specifically, the weighted disaster node-based topological network is a network model that integrates the spatial distribution of nodes and the energy conduction intensity. The dynamic weight matrix marks the connection intensity between nodes in the form of edge weights, realizing the simulatable expression of the damage propagation path. This network can characterize the interaction relationship between multiple damaged areas in the dam and provide structured input data for subsequent gravitational field simulation.

[0058] Specifically, a dynamically weighted disaster topology network is constructed through density clustering and spatial dissection. Its principle is based on screening key damage core nodes and dynamically quantifying the interaction intensity with the energy gradient. The technical effect is reflected in improving the identification accuracy of the spatial distribution of damaged nodes, enhancing the expression flexibility of complex structures through tetrahedral grid refinement, accurately depicting the non-uniformity of the energy conduction path using the dynamic weight matrix, providing a network topology basis with clear physical meaning for subsequent calculation of the probability of chain reaction, and thus improving the authenticity and reliability of the dam damage evolution simulation.

[0059] Optionally, the simulated energy conduction path includes:

[0060] Input the dynamic weight matrix of the topological network into the gravitational field model to calculate the potential energy field matrix between nodes;

[0061] Specifically, the gravitational field model is a mechanical simulation method based on the classical Newton's law of gravitation and topological weights. It maps the connection strength between nodes in the dynamic weight matrix into a gravitational interaction relationship. The potential energy field matrix is a mathematical representation matrix of the gravitational potential energy between nodes, which is obtained by calculating the spatial relationship between the node mass attributes and topological weights. The element values reflect the interaction potential energy strength between adjacent nodes during the energy conduction process, and are used to quantitatively describe the mechanical correlation between the internal damage areas of the dam.

[0062] Among them, the node mass is the integral value of the strain energy density of the corresponding grid cell, and the gravitational constant is adjusted according to the material Poisson's ratio , where is the material elastic modulus, is the Poisson's ratio; the weight matrix changes with the time step according to , where , is the aging coefficient, which is set according to the concrete grade. For example, for C30 concrete, = 0.05; the iteration termination condition is that the standard deviation of the path energy is less than 0.1 times the maximum potential energy peak value or the maximum step size condition is met.

[0063] Using the potential energy field matrix and the node set, an improved Markov chain Monte Carlo algorithm is used to simulate the energy conduction path, and the set of chain reaction probability values between adjacent nodes is output and stored in the dynamic risk probability database.

[0064] Specifically, the improved Markov chain Monte Carlo algorithm is a random path simulation method that combines topological weights and potential energy fields. It generates a state transition sequence through iteration to simulate the energy conduction process between nodes. The improvement points include introducing the potential energy gradient as a transfer probability correction factor and a dynamic weight decay mechanism. The set of chain reaction probability values is a multi-dimensional data set that records the energy conduction probability between adjacent nodes, and the dynamic risk probability database is a structured storage system that stores the risk probability evolution at different time steps and supports real-time query and update.

[0065] Among them, the improvements in the improved Markov chain Monte Carlo algorithm include: using the node potential energy difference as a constraint condition to correct the transfer probability, and setting the dynamic adjustment rule of the annealing coefficient ; setting the iteration termination condition as the potential energy convergence degree ≤ 0.01% for 100 consecutive simulations or the maximum step size reaches , The value is related to the fatigue characteristics of the material. For concrete,

[0066] The transition probability formula is corrected to:

[0067] ,

[0068] wherein, is the probability value of transferring from node to node . is the dynamic weight from node to node . is the annealing coefficient, is the potential energy difference from node to node . means normalizing the transition probabilities of all adjacent nodes of node to ensure that the sum of probabilities is 1. Its physical meaning is to ensure that the energy completes the conduction path selection within the neighborhood of node .

[0069] Specifically, the mechanical interaction relationship between nodes is quantified through a gravitational field model, and the random path of energy conduction is simulated based on an improved Monte Carlo method. Its principle is to reveal the mechanical mechanism of damage propagation by using the potential energy calculation driven by physics, and to capture the uncertainty of the conduction path by combining probability simulation. The technical effect is reflected in the synergistic drive of the dynamic weight and the potential energy gradient in the path simulation process, improving the physical rationality and computational efficiency of the prediction of the damage chain reaction. The generated set of chain reaction probability values can accurately reflect the spatio-temporal correlation of risk propagation among multiple nodes, providing high-confidence dynamic risk data support for the subsequent generation of risk heat maps and safety warnings.

[0070] Optionally, the spatio-temporal folding mapping operation includes:

[0071] Framing the multi-dimensional damage feature matrix according to the time series, and constructing a four-dimensional spatio-temporal hypercube in combination with the spatio-temporal distribution of the set of chain reaction probability values;

[0072] Specifically, the spatio-temporal folding mapping operation is a method for spatio-temporal correlation analysis of multi-dimensional data, which refers to a method of analyzing the cross-scale correlation by projecting the time and space dimensions onto the same hypercube coordinate system and using topological equivalent transformation. The mathematical implementation is the Hilbert space embedding of a four-dimensional tensor; framing the multi-dimensional damage feature matrix according to the time series and fusing the spatial distribution structure of the chain reaction probability values to construct a four-dimensional spatio-temporal hypercube. The four-dimensional spatio-temporal hypercube is a high-dimensional data structure formed with the three-dimensional space coordinates and the time axis as the basis, which can record the spatial evolution trajectory and probability coupling relationship of the damage features in different time slices simultaneously.

[0073] Performing discrete cosine transformation on the carbonization depth axis of the four-dimensional space-time hypercube to generate a frequency domain projection subspace;

[0074] Specifically, discrete cosine transform is a signal frequency domain processing technology. Discrete cosine transform of the carbonization depth axis means converting the time-varying sequence of carbonization depth into a frequency domain projection subspace. By extracting the characteristics of different frequency components, the periodic evolution law of the carbonization rate is revealed. The frequency domain projection subspace is the main frequency energy distribution matrix of the carbonization process, which can distinguish the spectral differences between normal carbonization and abnormal accelerated carbonization.

[0075] Performing second-order difference calculation on the macroscopic displacement cumulant axis of the four-dimensional space-time hypercube to extract frequency-domain-time-domain features;

[0076] Specifically, the second-order difference calculation is a method for extracting dynamic features of time series. The second-order difference calculation of the macro-displacement cumulative axis refers to quantifying the acceleration characteristics of the displacement change through the quadratic difference of adjacent time points. The frequency domain-time domain feature is a multidimensional feature set that integrates the frequency domain principal component and the time domain acceleration, which characterizes the inertial trend and sudden change fluctuation characteristics of the macro-displacement change.

[0077] The coupling coefficient tensor of the bidirectional diffraction equation is solved to generate the relationship curve between the carbonization depth change rate and the macroscopic displacement, and the carbonization-deformation coupling prediction curve is obtained.

[0078] Specifically, the two-way diffraction equation is a partial differential equation that couples carbonization and displacement variables. Solving the coupling coefficient tensor requires a set of matrix equations constructed by jointly constructing the frequency domain projection subspace and the second-order difference characteristics. The resulting carbonization-deformation coupling prediction curve is a dynamic correlation function of the daily change rate of carbonization depth and the accumulated displacement, which reflects the mutual influence mechanism between the two through the two-way action term.

[0079] in, , , where is the daily change rate of carbonization depth, is the accumulated macroscopic displacement, is the carbonization diffusion coefficient tensor, which is associated with the strain energy density field, is the coupling coefficient, which is calibrated by finite element simulation. The solution of the coupling coefficient is firstly normalized by orthogonalization of the time domain characteristic matrix and the frequency domain projection matrix, and then the least squares method is used to fit the displacement-carbonization coupling residual function.

[0080] Specifically, a bidirectional coupling model of carbonization and deformation is established through space-time folding mapping and multi-domain feature fusion. The principle is to decompose the four-dimensional space-time data into frequency domain-time domain features and solve the bidirectional action equation in parallel. The technical effect is reflected in breaking through the limitations of traditional single variable analysis, significantly enhancing the dynamic correlation modeling ability of microscopic carbonization process and macroscopic deformation, and accurately predicting the damage co-evolution trend through quantitative bidirectional coupling mechanism, providing multi-dimensional physical correlation basis for the full life cycle assessment of dam structure safety.

[0081] Optionally, the calculation of the damage energy exchange rate includes:

[0082] Extracting the instantaneous slope sequence of each monitoring point along the carbonization-deformation coupling prediction curve;

[0083] Specifically, the instantaneous slope sequence is a set of real-time change rates at each monitoring point of the carbonization-deformation coupling prediction curve, which is obtained by calculating the difference between adjacent time points on the curve, reflecting the dynamic relationship between the local deformation energy transfer rate and the carbonization process. This sequence can quantify the damage energy efficiency conversion rate at different locations at a specific moment, providing transient dynamic characteristics for energy exchange rate calculation.

[0084] Aligning the instantaneous slope sequence with the chain reaction probability value set according to topological network node coding;

[0085] Specifically, the topological network node coding alignment is an operation that matches the instantaneous slope sequence with the chain reaction probability value according to the spatial coordinate and topological structure mapping relationship. It is necessary to achieve accurate correspondence between the time and space dimensions through the node unique identifier to ensure the homology of the deformation energy rate and the risk probability data, and establish the spatiotemporal consistency at the data level for subsequent fusion analysis.

[0086] The aligned encoded data is fused with the ambient temperature and humidity time series data based on the Lagrange interpolation algorithm to generate a dynamic exchange rate correction factor;

[0087] Specifically, the Lagrange interpolation algorithm is a function interpolation method based on polynomial fitting. It generates a dynamic exchange rate correction factor by interpolating and fusing the aligned data with the ambient temperature and humidity time series data at the same time node. This factor quantifies the modulation effect of temperature and humidity changes on energy conduction efficiency and weakens the interference of environmental noise on the calculation of energy exchange rates.

[0088] The noise interference is eliminated by principal component analysis and the damage energy exchange rate is output.

[0089] Specifically, the principal component analysis method is a dimensionality reduction and denoising method that extracts the main components of data through orthogonal transformation. It decomposes the covariance matrix of the fused multi-source data, retains a few principal components that represent the essential characteristics of the energy exchange rate, eliminates sensor noise and redundant coupling features, and outputs an injury energy exchange rate index with a high signal-to-noise ratio, accurately representing the energy conduction efficiency per unit time at each monitoring point.

[0090] Specifically, the calculation accuracy of the injury energy exchange rate is improved by spatio-temporal alignment of multi-source data and environmental parameter correction. The principle lies in combining the dynamic characteristics of the deformation slope and the topological network risk probability information, fusing the influence of environmental factors through an interpolation algorithm, and refining the core features using principal component analysis. The technical effect is reflected in enhancing the quantitative representation ability of the injury energy conduction process, reducing the misguidance of the external environment disturbance and data noise to the evaluation model, ensuring that the generated energy exchange rate data can reliably support the calculation of the subsequent risk coupling heat map, and improving the robustness and accuracy of the dynamic assessment of the dam safety risk.

[0091] Optionally, the generation of the standardized risk coupling coefficient heat map includes:

[0092] Mapping the injury energy exchange rate and the chain reaction probability value to the grid vertices of the topological network;

[0093] Specifically, the topological grid vertices are the set of connection points of the three-dimensional space model in the disaster node-based topological network. Mapping to the grid vertices of the topological network means assigning the injury energy exchange rate and the chain reaction probability value as dynamic attributes to the grid nodes, so that each vertex carries both energy conduction efficiency and risk diffusion probability data. This operation establishes a strong correlation between data and spatial positions, providing a structured input for subsequent matrix operations.

[0094] Constructing a three-dimensional covariance Gram matrix and strengthening the non-linear coupling effect through a hyperbolic tangent activation function;

[0095] Specifically, the three-dimensional covariance Gram matrix is a high-dimensional extended matrix of the multi-variable covariance relationship, generated by calculating the cross-covariance between the injury energy exchange rate, the chain reaction probability value, and the spatio-temporal coordinates; the hyperbolic tangent activation function is a non-linear function with an S-shaped curve, which compresses the input to the [-1, 1] interval and enhances the non-linear correlation, strengthening the complex coupling effect between the energy exchange rate and the probability value, and significantly improving the dynamic interaction feature expression ability between network nodes.

[0096] Using the t-SNE algorithm in manifold learning to reduce the dimension to a two-dimensional plane;

[0097] Specifically, the t-SNE algorithm in manifold learning is a data dimensionality reduction method based on probability distribution. By preserving the probability distribution characteristics of local similarity in high-dimensional space, it projects the three-dimensional covariance matrix data onto a two-dimensional plane. This algorithm can effectively distinguish the aggregation pattern and abnormal features of energy conduction during the dimensionality reduction process, and generate an intuitively analyzable risk distribution base image.

[0098] Dynamically superimpose the risk coefficient distribution according to the time label of the spatio-temporal hypercube, and output a standardized heat map.

[0099] Specifically, the time label of the spatio-temporal hypercube is an index identifier that records the monitoring moment in the four-dimensional data structure. Dynamically superimposing the risk coefficient distribution means weighted fusion of the heat layers under different time slices. The output standardized heat map eliminates the dimension difference through normalization processing, dynamically shows the gradient distribution of the overall risk coupling strength over time, and realizes the spatio-temporal continuous expression of risk visualization.

[0100] Specifically, a risk heat map is generated through multi-dimensional matrix construction and non-linear enhanced dimensionality reduction. Its principle lies in fusing the spatial covariance relationship between the energy exchange rate and the probability value, and reflecting the risk evolution process through dynamic time superposition. The technical effect is to break through the limitations of traditional static heat maps, enhance the quantitative characterization ability of non-linear coupling relationships, intuitively display the spatio-temporal aggregation characteristics and diffusion trends of damage propagation, provide a high-resolution visualization tool for potential hazard identification, and at the same time support the dynamic monitoring and decision-making intervention of risk evolution.

[0101] Optionally, the multi-axis domain feature distillation includes:

[0102] Calculate the directional differential operator along the heat gradient axis of the standardized risk coupling coefficient heat map, and extract the geometric feature vector of the gravitational field intensity mutation region;

[0103] Specifically, the directional differential operator is a mathematical tool for calculating the local change intensity along the heat gradient axis. Calculating the directional differential operator along the heat gradient axis means performing partial derivative operations on the heat map along the spatial coordinate axes to extract the geometric feature vector of the gravitational field intensity mutation region. This vector contains geometric attributes such as the gradient amplitude and direction curvature of the mutation region, and can quantitatively describe the distortion degree and spatial distribution pattern of the internal stress field of the dam.

[0104] Perform a fast Fourier transform on the heat frames of consecutive time periods, and calculate the power spectrum entropy as the seepage phase spectrum entropy;

[0105] Specifically, the fast Fourier transform is a classic algorithm for converting time-domain signals to the frequency domain. Performing the fast Fourier transform on thermal frames in continuous time periods means converting the local thermal distribution data of the time series into spectral components; the power spectrum entropy is an entropy value index obtained by calculating the distribution chaos degree of spectral energy. As the seepage phase spectrum entropy, it directly reflects the complexity and randomness of the seepage process in the frequency domain. A high entropy value indicates disordered seepage paths or abnormal seepage behavior.

[0106] Among them, the time-frequency transformation parameters are: the FFT window length is 2048 points, the Hanning window is used, and the overlap rate is 50%; the frequency bands are divided into 16 sub-bands according to the critical frequency of concrete permeability, and the critical frequency is 0.1 - 10 Hz; the power spectrum entropy , where is the proportion of the energy of the th sub-band, normalized between 0 and 1; the safety domain is determined as when the power spectrum entropy > 0.85 and the duration > 1 hour, triggering an abnormal seepage alarm.

[0107] Combining the energy convergence trajectory of the spatio-temporal hypercube to fit the Lyapunov exponent to generate the approaching energy convergence degree;

[0108] Output a set of hidden danger quantification indicators including mutation gradient, spectrum entropy, and convergence degree.

[0109] Specifically, the energy convergence trajectory is the historical evolution path of the energy density in the spatio-temporal hypercube. Combining the energy convergence trajectory to fit the Lyapunov exponent means calculating the divergence characteristic index by statistically analyzing the attenuation or accumulation rate of energy over time. The generated approaching energy convergence degree is used to evaluate the stability of the energy conduction process in the dam damage area. The Lyapunov exponent is a dynamic index that quantifies the sensitivity of the system to initial conditions. A negative value indicates that the energy tends to converge stably, and a positive value indicates a potential unstable state.

[0110] Among them, based on the energy convergence trajectory data of the spatio-temporal hypercube, the Wolf algorithm is used to fit the maximum Lyapunov exponent, the data sampling interval Δt = 1h, and the embedding dimension d = 3.

[0111] Specifically, hidden danger quantification indicators are extracted by distilling the multi-axis domain characteristics of the thermal map. Its principle is to analyze risk characteristics from three dimensions: geometric mutation, frequency domain chaos degree, and dynamic stability. The technical effect is to enhance the comprehensiveness and physical interpretability of the hidden danger indicators by integrating geometric, spectral, and dynamic methods. By quantifying the mutation gradient, the stress concentration area is located, the seepage anomaly degree is evaluated using the spectrum entropy, and the system stability is predicted by combining the energy convergence degree, providing a multi-dimensional evaluation basis for the refined classification and precise intervention of dam safety risks.

[0112] Optionally, the dynamic comparison includes:

[0113] Input the set of quantified hidden danger indicators into the fuzzy Petri net inference engine in the knowledge base;

[0114] Match the critical interval of the elastic modulus of the gravitational field strength and the seepage phase safety domain in the safety envelope function;

[0115] Specifically, the fuzzy Petri net inference engine is an intelligent decision-making tool that combines fuzzy logic and the dynamic deduction rules of Petri nets. Inputting the set of quantified hidden danger indicators into this engine means quantifying the degree of deviation of the indicators from the safety threshold by defining fuzzy membership functions, triggering the preset fuzzy inference rule chain in the knowledge base. Matching the critical interval of the elastic modulus of the gravitational field strength and the seepage phase safety domain in the safety envelope function requires comparing the membership relationship between the measured indicators and the envelope boundary to confirm whether each indicator exceeds the safety threshold of material mechanical properties or seepage dynamics. Among them, the critical interval of the elastic modulus of the gravitational field strength is the allowable stress fluctuation range set according to material performance experiments, and the seepage phase safety domain is the allowable operation domain of seepage velocity and flow direction angle.

[0116] Among them, for the critical interval of the elastic modulus of the safety envelope function, the yield strength threshold is calibrated through the tensile-compression test of dam materials, and the elastic modulus fluctuation range is generated by combining the Monte Carlo sampling method as the safety domain boundary, and the fluctuation range is ±10% of the reference value. For example, 100 groups of concrete core samples of the dam body are collected, the elastic modulus is tested according to ASTM C469 standard, the dynamic fluctuation threshold is calculated, and the dynamic fluctuation threshold is the standard deviation of ±3 times the reference value. After excluding outliers, a confidence interval of ±10% is taken; verified by finite element simulation: when the local elastic modulus value exceeds the dynamic fluctuation threshold, the displacement increment error is greater than 15%.

[0117] When any indicator breaks through the envelope boundary, trigger the Monte Carlo tree search protocol of the three-dimensional decision engine;

[0118] Based on the optimal convergence solution of the energy conduction path and the coupling prediction curve, output the reinforcement strength vector diagram and the flood discharge scheduling sequence.

[0119] Specifically, the Monte Carlo tree search protocol of the three-dimensional decision engine is a random search algorithm based on a tree structure. After triggering this protocol, the optimal response strategy is calculated by simulating countless potential decision paths. The optimal convergence solution is the comprehensive optimal solution that satisfies the low dissipation of the energy conduction path and the smoothness of the carbonization-deformation coupling prediction curve. The reinforcement strength vector diagram is a three-dimensional spatial distribution diagram representing the strength and direction of the reinforcement materials applied in different regions, and the flood discharge scheduling sequence is an optimized operation instruction set that controls the flood discharge flow rate and time interval of the reservoir to ensure the reduction of water pressure load before the risk is eliminated.

[0120] Specifically, the hidden danger warning and the dynamic generation of the disposal plan are realized through fuzzy reasoning and Monte Carlo tree search. The principle lies in using the fuzzy Petri net to capture the fuzzy state transition of the index out-of-bounds, and combining the tree search algorithm to quickly plan the optimal response strategy. The technical effect is reflected in breaking through the single judgment mode of the traditional threshold alarm, realizing the intelligent reasoning and hierarchical response of the multi-index associated risks, and the generated reinforcement vector diagram and flood discharge sequence accurately matching the current actual damage state and environmental load of the dam, significantly improving the scientificity and implementation efficiency of the safety disposal measures. Embodiment 1:

[0121] To verify the feasibility and effectiveness of the present invention in dam safety monitoring, a certain large concrete gravity dam is taken as an example for implementation. The dam is 185 meters high, with a total reservoir capacity of 12 billion cubic meters, equipped with an acoustic emission sensor array and an infrared thermal imaging system. The acoustic emission sensor array has 64 channels, and the resolution of the infrared thermal imaging system is 0.05 °C, constructing a digital monitoring network covering the surface and internal structure of the dam body.

[0122] During the flood season in 2023, the system collected the acoustic emission signals on the dam surface and the internal infrared thermal image data in real time. The sampling rate of the acoustic emission signals was 1 MHz, and the internal infrared thermal image data was a full-dam scan every 30 minutes. At 10:23 on July 15th, a sudden anomaly in the power spectrum of the acoustic emission signal was detected in the C12 area of the dam section. The characteristic frequency jumped from 25 kHz to 82 kHz. At the same time, the infrared thermal image showed that there was a low-temperature gradient anomaly area with a diameter of 3.2 m in this area, ΔT = 1.8 °C / m.

[0123] Through cross-modal convolution kernel fusion, the acoustic emission signal is decomposed into 8 frequency bands by wavelet packet decomposition, generating an energy distribution tensor of 64×64×8; after the infrared thermal image is processed by morphological gradient, the detected gradient peak reaches 4.8 °C / m². The finally generated damage characteristic matrix shows that the strain energy density in the C12 area reaches 5.7 kJ / m³, and the average value < 1.2 kJ / m³, which is determined as a Class III damage area.

[0124] Based on the density peak clustering algorithm, 35 core nodes are selected to form a disaster network. Delaunay hierarchical triangulation generates 412 tetrahedral elements, and the energy gradient between adjacent elements is calculated. The energy gradient of the element pair (C12-24, C12-25) is 4.7 kJ / m 4 and the energy gradient of the element pair (C12-25, C12-26) is 8.3 kJ / m 4, critical jump. The node weights are calculated by the gravitational field model, the gravitational constant G is 0.12MN·m / t², the concrete parameter is C30, and after 20,000 Monte Carlo simulations, the chain reaction probability values ​​of the C12 node group are output. The chain reaction probability of the associated nodes (C12-25, C13-03) is 0.78, the chain reaction probability of the associated nodes (C12-25, C11-17) is 0.65, and the chain reaction probability of the associated nodes (C12-25, spillway base F09) is 0.43.

[0125] The hypercube dimension generated by the space-time folding mapping has a space-time axis of 30 days, 72 time slices, and a coordinate accuracy of 0.1m; the deformation monitoring point AX05 shows a cumulative displacement of 12.7mm; the solution of the two-way diffraction equation shows that there is a strong coupling between the daily growth rate of carbonization depth and the displacement acceleration stage, such as Figure 2 As shown, the prediction curve shows that if the carbonization rate reaches 2.5 μm / d, the displacement will exceed the threshold after 28 days, and the fitting error is ±0.11 mm.

[0126] Through Gram matrix calculation, a standardized risk heat map is generated, which shows that the risk coefficient of the C12 core area is 0.93, the risk coefficient of the spillway base F09 is 0.68, and the risk coefficient of the left bank drainage corridor L24 is 0.55. The multi-axis domain feature distillation results are that the gravitational field gradient mutation is 0.43kN / m²·m, the threshold is greater than 0.33kN / m²·m; the seepage entropy value is 0.89, the safety domain is less than or equal to 0.8, the Lyapunov index is 0.12, and the steady state should be less than 0.

[0127] The comparison of the safety envelope function shows that the seepage entropy value is 0.89, which breaks the threshold and triggers the three-dimensional decision engine. The Monte Carlo tree search outputs the micro-repair plan: the coordinate set is the position of {C12-25(34.72M,121.85M,EL.152)}, and the epoxy mortar injection volume is 2.4m³; flood discharge scheduling optimization: the scheduling sequence is [0-2h: 500m³ / s to 2-6h:800m³ / s], and the gate opening G05 is controlled to be 62% and G07 is controlled to be 58%; reinforcement vector plan: the density of the anchor group is 3.2 roots / m² in the C12 area, the vector angle is 45 degrees, and the density in the F09 area is 2.1 roots / m², and the vector angle is 32 degrees.

[0128] After 48 hours of implementing the treatment plan, the strain energy density of the C12 area was re-measured and dropped to 1.8kJ / m³, a decrease of 68%; the seepage entropy dropped back to 0.7, within the qualified threshold; the displacement rate dropped to 0.13mm / d, a decrease of 59%. This embodiment verifies the dynamic monitoring capability of the system in a complex environment. Through multi-source data fusion and intelligent decision-making models, closed-loop management from early detection of damage to precise treatment is achieved, providing innovative solutions for the safe operation of the dam.

[0129] Based on the same inventive concept, such as Figure 3 shown, the present invention also provides a dam safety monitoring system based on digital twin, and the system includes:

[0130] A feature acquisition module, configured to obtain the acoustic emission signal distribution map on the dam surface and the internal infrared thermal image matrix, and generate a multi-dimensional damage feature matrix including the strain energy density field through cross-modal convolution kernel fusion calculation;

[0131] A disaster network construction module, configured to establish a disaster node-based topological network based on the spatial coordinates of the multi-dimensional damage feature matrix;

[0132] A reaction probability calculation module, configured to input the node weights in the topological network into a gravitational field strength calculation model, simulate the energy conduction path, and output a set of chain reaction probability values between adjacent nodes in the path;

[0133] A carbonization-deformation coupling prediction curve generation module, configured to perform spatio-temporal folding mapping operation on the multi-dimensional damage feature matrix according to the set of chain reaction probability values, establish a bidirectional diffraction equation between the daily change rate sequence of the microscopic carbonization depth and the cumulative macroscopic displacement, and generate a carbonization-deformation coupling prediction curve;

[0134] A risk coupling coefficient heat map generation module, configured to calculate the damage energy exchange rate at each monitoring point based on the carbonization-deformation coupling prediction curve, perform non-linear Gram matrix multiplication on the energy exchange rate and the chain reaction probability value at the corresponding position, and generate a standardized risk coupling coefficient heat map;

[0135] A hidden danger quantification index extraction module, configured to perform multi-axis domain feature distillation on the risk coupling coefficient heat map, extract hidden danger quantification indexes including the mutation gradient of the gravitational field strength, the seepage phase spectrum entropy, and the approaching energy convergence degree, and perform dynamic comparison with the safety envelope function in the knowledge base;

[0136] A solution generation module, configured to generate a disposal solution including a microscopic repair coordinate set, a flood discharge path rescheduling sequence, and a reinforcement strength vector diagram when any hidden danger quantification index breaks through the preset envelope boundary.

[0137] It should be noted that the electrical connections between the above-mentioned various units do not necessarily represent direct connections of the circuits. Indirect connection methods, as long as the purpose of the present invention is achieved, can be applied to the embodiments of the present invention. The above are only exemplary embodiments of the present invention, and the scope of the present invention cannot be limited thereby.

[0138] That is, any equivalent changes and modifications made in accordance with the teachings of the present invention still fall within the scope covered by the present invention. Those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and the disclosure of the practice. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include the common general knowledge or conventional technical means in the technical field not recorded in the present invention.

Claims

1. A dam safety monitoring method based on digital twins, characterized in that: The method comprises: Obtain the distribution spectrum of acoustic emission signals on the dam surface and the internal infrared thermal image matrix, and generate a multi-dimensional damage feature matrix containing the strain energy density field through cross-modal convolution kernel fusion calculation; Based on the spatial coordinates of the multi-dimensional damage feature matrix, a disaster node topological network is established; Input the node weights in the topological network into the gravitational field strength calculation model, simulate the energy conduction path, and output a set of chain reaction probability values ​​between adjacent nodes in the path; According to the chain reaction probability value set, a time-space folding mapping operation is performed on the multi-dimensional damage characteristic matrix, a bidirectional diffraction equation between the microscopic carbonization depth daily change rate sequence and the macroscopic displacement accumulation is established, and a carbonization-deformation coupling prediction curve is generated; Based on the carbonization-deformation coupling prediction curve, the damage energy exchange rate at each monitoring point is calculated, and the energy exchange rate is multiplied by the chain reaction probability value at the corresponding position by nonlinear Gram matrix to generate a standardized risk coupling coefficient heat map; Perform multi-axis domain feature distillation on the risk coupling coefficient heat map to extract hidden danger quantitative indicators including gravitational field intensity mutation gradient, percolation phase spectrum entropy, and convergence degree of approaching energy, and dynamically compare them with the safety envelope function in the knowledge base; When any quantitative indicator of hidden danger exceeds the preset envelope boundary, a disposal plan is generated including a micro-repair coordinate set, a flood discharge path rescheduling sequence, and a reinforcement strength vector diagram.

2. A dam safety monitoring method based on digital twins according to claim 1, characterized in that: The cross-modal convolution kernel fusion calculation includes: Performing wavelet packet decomposition on the acoustic emission signal distribution spectrum to generate a multi-band energy distribution tensor; Performing edge enhancement based on a morphological gradient operator on the internal infrared thermal image matrix, and outputting a spatial gradient matrix of the thermal conduction abnormality area; The multi-band energy distribution tensor and the spatial gradient matrix are input into a dual-channel residual network for cross-modal feature fusion, and the fusion weights are optimized through the strain energy density field loss function to generate the multi-dimensional damage feature matrix.

3. A dam safety monitoring method based on digital twins according to claim 2, characterized in that: The establishment of the disaster node topology network includes: Based on the strain energy density field distribution of the multi-dimensional damage characteristic matrix, a density peak clustering algorithm is used to screen the potential disaster node set; Perform Delaunay hierarchical decomposition on the node set to construct a three-dimensional space grid model and divide it into multiple grid units; Calculate the energy gradient change rate between adjacent grid cells and generate a dynamic weight matrix; The dynamic weight matrix is ​​used to output a disaster node topological network with weights.

4. A dam safety monitoring method based on digital twins according to claim 3, characterized in that: The simulated energy conduction path includes: Inputting the dynamic weight matrix of the topological network into the gravitational field model to calculate the potential field matrix between nodes; The potential energy field matrix and the node set are used to simulate the energy conduction path using an improved Markov chain Monte Carlo algorithm, and a chain reaction probability value set between adjacent nodes is output and stored in a dynamic risk probability database.

5. A dam safety monitoring method based on digital twins according to claim 4, characterized in that: The space-time folding mapping operation includes: Divide the multi-dimensional damage feature matrix into frames according to time series, and construct a four-dimensional space-time hypercube by combining the space-time distribution of the chain reaction probability value set; Performing discrete cosine transformation on the carbonization depth axis of the four-dimensional space-time hypercube to generate a frequency domain projection subspace; Performing second-order difference calculation on the macroscopic displacement cumulant axis of the four-dimensional space-time hypercube to extract frequency-domain-time-domain features; The coupling coefficient tensor of the bidirectional diffraction equation is solved to generate the relationship curve between the carbonization depth change rate and the macroscopic displacement, and the carbonization-deformation coupling prediction curve is obtained.

6. A dam safety monitoring method based on digital twins according to claim 5, characterized in that: The calculation of the damage energy exchange rate includes: Extracting the instantaneous slope sequence of each monitoring point along the carbonization-deformation coupling prediction curve; Aligning the instantaneous slope sequence with the chain reaction probability value set according to topological network node coding; The aligned encoded data is fused with the ambient temperature and humidity time series data based on the Lagrange interpolation algorithm to generate a dynamic exchange rate correction factor; The noise interference is eliminated by principal component analysis and the damage energy exchange rate is output.

7. A dam safety monitoring method based on digital twins according to claim 6, characterized in that: The generating of the standardized risk coupling coefficient heat map comprises: Mapping the damage energy exchange rate and the chain reaction probability value to the grid vertices of the topological network; Construct a three-dimensional covariance Gram matrix and enhance the nonlinear coupling effect through the hyperbolic tangent activation function; Use the t-SNE algorithm in manifold learning to reduce the dimension to a two-dimensional plane; The risk coefficient distribution is dynamically superimposed according to the time labels of the space-time hypercube, and a standardized heat map is output.

8. A dam safety monitoring method based on digital twins according to claim 7, characterized in that: The multi-axis domain feature distillation comprises: Calculating a directional differential operator along the thermal gradient axis of the standardized risk coupling coefficient thermal map to extract geometric characteristic vectors of the region where the gravitational field intensity changes suddenly; Perform fast Fourier transform on the thermal frames of continuous time periods and calculate the power spectrum entropy as the percolation phase spectrum entropy; Fitting the Lyapunov exponent in combination with the energy convergence trajectory of the space-time hypercube to generate an approximate energy convergence degree; The output includes a set of hidden danger quantitative indicators including mutation gradient, spectral entropy and convergence degree.

9. A dam safety monitoring method based on digital twins according to claim 8, characterized in that: The dynamic comparison includes: Inputting the hidden danger quantitative indicator set into the fuzzy Petri net reasoning engine in the knowledge base; Match the critical interval of elastic modulus of gravitational field strength and the safe region of seepage phase in the safety envelope function; When any indicator breaks through the envelope boundary, the Monte Carlo tree search protocol of the three-dimensional decision engine is triggered; Based on the optimal convergence solution of the energy conduction path and the coupling prediction curve, a reinforcement strength vector diagram and a flood discharge scheduling sequence are output.

10. A dam safety monitoring system based on digital twins, applied to the dam safety monitoring method based on digital twins according to any one of claims 1 to 9, characterized in that: The system comprises: The feature acquisition module is used to obtain the distribution map of acoustic emission signals on the dam surface and the internal infrared thermal image matrix, and generate a multi-dimensional damage feature matrix including the strain energy density field through cross-modal convolution kernel fusion calculation; A disaster network construction module is used to establish a disaster node topological network based on the spatial coordinates of the multi-dimensional damage feature matrix; A reaction probability calculation module, used to input the node weights in the topological network into the gravitational field strength calculation model, simulate the energy conduction path, and output a set of chain reaction probability values ​​between adjacent nodes in the path; A carbonization-deformation coupling prediction curve generation module is used to perform a spatiotemporal folding mapping operation on the multi-dimensional damage feature matrix according to the chain reaction probability value set, establish a bidirectional diffraction equation between the microscopic carbonization depth daily change rate sequence and the macroscopic displacement accumulation, and generate a carbonization-deformation coupling prediction curve; A risk coupling coefficient heat map generation module is used to calculate the damage energy exchange rate at each monitoring point based on the carbonization-deformation coupling prediction curve, perform nonlinear Gram matrix multiplication on the energy exchange rate and the chain reaction probability value at the corresponding position, and generate a standardized risk coupling coefficient heat map; A hidden danger quantitative index extraction module is used to implement multi-axis domain feature distillation on the risk coupling coefficient heat map, extract hidden danger quantitative indicators including gravitational field intensity mutation gradient, percolation phase spectrum entropy, and approximate energy convergence, and dynamically compare them with the safety envelope function in the knowledge base; The solution generation module is used to generate a treatment plan including a micro-repair coordinate set, a flood discharge path rescheduling sequence, and a reinforcement strength vector diagram when any quantitative indicator of hidden danger exceeds the preset envelope boundary.

Images