Targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing

Abstract

The present invention relates to the technical field of underground structure engineering repair, and discloses a method for targeted diagnosis and treatment of underground structure cracks in hot and humid areas combined with non-destructive testing. The method first collects multi-source signals such as three-dimensional deformation data and crack distribution parameters of the underground structure to construct a structural damage characteristic tensor; then performs dynamic topological quantification processing on it, establishes a multi-modal coupling network to evaluate the cracking state, and generates targeted repair decision parameters; then constructs a structural damage constraint function, calculates the dynamic equilibrium boundary of the cracked area; finally, optimizes the repair parameters based on the damage characteristic tensor and the equilibrium boundary, and generates a targeted repair execution sequence. In addition, the repair process is dynamically adjusted through closed-loop control. The present invention can accurately detect and target the problem of underground structure cracks in hot and humid areas, improve the pertinence and effectiveness of repairs, ensure the safe and stable operation of underground structures, and has significant practical value.

Description

Technical Field

[0001] The present invention relates to the technical field of underground structure engineering repair, and in particular to a targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing. Background Art

[0002] In modern urban construction, underground structures are widely used in areas such as subways, underground parking lots, underground shopping malls, and various other municipal infrastructure. Their safety and stability are directly related to the normal operation of cities and the safety of people's lives and property. In hot and humid regions, due to their unique climate, underground structures face even more severe challenges.

[0003] Hot and humid regions experience year-round high temperatures and high humidity, accelerating the aging process of underground structural materials. For example, under the long-term effects of heat and humidity, cement hydration products in the concrete react with carbon dioxide in the air to form carbonation structures, reducing the concrete's alkalinity and damaging the passivation film on the steel bars, leading to steel corrosion. Corroded steel bars expand in volume, exerting compressive pressure on the surrounding concrete, which can cause cracking. Furthermore, hot and humid environments promote the growth and reproduction of microorganisms, whose metabolic products can corrode the concrete, further weakening its structural performance.

[0004] Groundwater levels are typically higher in humid and hot regions, and underground structures are subject to long-term immersion and infiltration by groundwater. Groundwater contains various chemical components, such as sulfates and chlorides, which react with components in concrete to produce expansive products, causing cracking and damage to concrete structures. For example, sulfates react with calcium hydroxide in concrete to form gypsum, which in turn reacts with tricalcium aluminate in cement to form ettringite. The volume expansion of ettringite can cause cracks within concrete structures. Furthermore, groundwater infiltration can cause uneven settlement of underground structures, increasing the risk of structural cracking.

[0005] Traditional methods for detecting and repairing cracks in underground structures have numerous limitations in hot and humid regions. Regarding detection, commonly used visual inspection methods struggle to detect early-stage internal damage and fine cracks. While localized damage detection methods can obtain relatively accurate internal information, they can damage the structure, impacting its normal use. Furthermore, their detection efficiency is low, making them inefficient and unable to meet the needs of large-scale inspections. Regarding repair, traditional methods often lack specificity, often employing a unified repair plan that fails to fully consider the causes and extent of cracks in different locations, as well as the actual stress conditions of the structure. This can not only lead to poor repair results and an inability to effectively prevent further crack expansion, but can also result in a waste of resources.

[0006] Due to the lack of effective detection and repair methods, once cracking occurs in underground structures in hot and humid areas, if not addressed promptly, it will gradually worsen over time and, in severe cases, even lead to structural collapse. This not only affects the normal use of underground space and causes huge economic losses, but also may cause safety accidents and threaten people's lives. Therefore, it is urgent to develop a method that can accurately detect and target the diagnosis and treatment of cracking in underground structures in hot and humid areas. Summary of the Invention

[0007] The purpose of the present invention is to provide a targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing to solve the problems raised in the above background technology.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing, the method comprising:

[0009] Collect 3D deformation data, crack distribution parameters, ambient temperature and humidity signals, and material stress response signals of underground structures, and construct a structural damage characteristic tensor;

[0010] Performing dynamic topological quantization on the structural damage characteristic tensor to generate a damage evolution parameter matrix, and establishing a multimodal coupling network to evaluate the cracking state and generate targeted repair decision parameters;

[0011] Constructing a structural damage constraint function according to the targeted repair decision parameters, and calculating a dynamic equilibrium boundary of the cracked area;

[0012] Repair parameters are iteratively optimized based on the structural damage characteristic tensor and the dynamic equilibrium boundary to generate a targeted repair execution sequence.

[0013] Preferably, the collecting of three-dimensional deformation data, crack distribution parameters, environmental temperature and humidity signals, and material stress response signals of the underground structure, and constructing a structural damage characteristic tensor, includes:

[0014] Perform multi-source synchronous acquisition of 3D deformation data, crack distribution parameters, ambient temperature and humidity signals, and material stress response signals to generate a non-destructive testing raw data matrix;

[0015] Performing spatial normalization processing on the original data matrix to obtain a standardized data matrix, and performing modal separation on the standardized data matrix based on a tensor decomposition algorithm to generate initial components of damage characteristics;

[0016] Performing time domain alignment correction on the initial damage characteristic component, performing phase compensation by introducing an environmental coupling coefficient, and generating a corrected damage characteristic component;

[0017] Based on a multi-scale fusion algorithm, the corrected damage characteristic components and the material degradation matrix are correlated and analyzed to generate a damage correlation tensor. Multi-dimensional features are extracted from the damage correlation tensor to generate structural damage characteristic tensors under different working conditions.

[0018] Preferably, the dynamic topological quantization processing of the structural damage characteristic tensor to generate a damage evolution parameter matrix, and the establishment of a multimodal coupling network to evaluate the cracking state and generate targeted repair decision parameters include:

[0019] Performing regional block quantization on the structural damage feature tensor to generate damage distribution interval parameters, and performing dynamic threshold calibration on the structural damage feature based on the damage distribution interval parameters to generate a damage calibration matrix;

[0020] Performing multi-level division processing on the damage feature data in the structural damage feature tensor according to the damage calibration matrix to generate a block feature matrix;

[0021] Performing weighted topological processing on the block feature matrix and setting a dynamic evolution interval to generate a topological feature matrix;

[0022] The topological characteristic matrix is ​​subjected to environmental noise suppression by a nonlinear filter to generate a noise reduction characteristic matrix, and the noise reduction characteristic matrix is ​​subjected to tensor normalization conversion to generate a damage evolution parameter matrix;

[0023] Based on the damage evolution parameter matrix, a multimodal coupling network is constructed, and the cracking damage state of the underground structure is evaluated in real time to generate targeted repair decision parameters.

[0024] Preferably, the multimodal coupling network is constructed based on the damage evolution parameter matrix, and the cracking damage state of the underground structure is evaluated in real time to generate targeted repair decision parameters, including:

[0025] Dividing the damage evolution parameter matrix into a repair feature matrix and an evaluation feature matrix;

[0026] A repair decision network is constructed based on the repair feature matrix. The repair decision network includes an input layer, a hidden layer, and an output layer. The input layer contains four neurons corresponding to deformation, crack, temperature and humidity, and stress signals. The hidden layer uses a piecewise nonlinear activation function. The output layer generates a repair decision value to obtain the repair network output data.

[0027] Constructing a damage assessment network based on the assessment feature matrix, the damage assessment network includes an input layer, a hidden layer, and an output layer, wherein the input layer receives the output data of the repair network, the hidden layer uses a hyperbolic tangent activation function, and the output layer generates a damage quantization value to obtain the output data of the assessment network;

[0028] Jointly training the repair decision network and the damage assessment network to generate a multimodal coupled network structure, and performing residual function calculation and network robustness analysis on the multimodal coupled network structure to generate a network evaluation result;

[0029] Based on the network evaluation results, the parameters of the repair decision network and the damage assessment network in the multimodal coupling network structure are dynamically updated, and the network weights are adjusted through the back propagation algorithm to generate targeted repair decision parameters.

[0030] Preferably, the step of constructing a structural damage constraint function based on the targeted repair decision parameters and calculating the dynamic equilibrium boundary of the cracked area includes:

[0031] Performing an asymmetric matrix construction on the targeted repair decision parameters and generating a damage state matrix through linear transformation;

[0032] Designing dynamic integral constraint terms based on the damage state matrix, performing piecewise weighting operations on damage variables to generate equilibrium constraint terms, and combining the damage state matrix and the equilibrium constraint terms to construct a structural stability function to generate a structural damage constraint function;

[0033] Calculating the spatial gradient of the structural damage constraint function to generate a gradient analysis result, and performing a comparison operation on the gradient analysis result and the modulus length of the damage vector to generate a damage constraint condition;

[0034] The repair strength interval of the cracked area is partitioned according to the damage constraint condition, and a dynamic balance coefficient is set to generate a constraint range. Boundary fitting is performed based on the constraint range to generate a dynamic balance boundary of the cracked area.

[0035] Preferably, the iterative optimization of repair parameters based on the structural damage characteristic tensor and the dynamic equilibrium boundary to generate a targeted repair execution sequence includes:

[0036] Constructing a repair weight matrix and a material compensation matrix according to the structural damage characteristic tensor, and substituting the repair weight matrix and the material compensation matrix into a dynamic optimization function to generate a repair objective function;

[0037] Applying material strength constraints and repair rate constraints to the repair objective function, and converting the dynamic equilibrium boundary into a gradient convergence constraint to generate an optimization constraint;

[0038] Inputting the repair objective function and the optimization constraints into a numerical iterative solver, calculating the repair parameters in a discrete space, and generating an initial sequence of repair parameters;

[0039] Adaptive step size adjustment is performed on the initial sequence of repair parameters, and stress constraint processing is performed in combination with the material properties of the underground structure to generate a targeted repair execution sequence.

[0040] Preferably, the method further comprises:

[0041] Performing process conversion on the targeted repair execution sequence to generate a construction control instruction, and performing error compensation on the construction control instruction to generate a repair execution signal;

[0042] Calculate the target repair strength based on the stability requirements of the underground structure, monitor the crack expansion rate in real time, and generate crack monitoring data;

[0043] Calculating the difference between the crack monitoring data and the target repair strength value to generate a repair deviation value, and comparing the repair deviation value with a preset repair threshold to generate a deviation determination result;

[0044] Dynamically correcting the repair parameters based on the deviation determination result to generate updated repair parameters, and performing construction sequence conversion on the updated repair parameters, converting the repair parameters into construction intensity values ​​through material property mapping to generate a new repair execution sequence;

[0045] The new repair execution sequence is input into the repair execution device for closed-loop control, and the construction process is dynamically adjusted to complete the targeted repair of underground structure cracks.

[0046] Preferably, performing weighted topological processing on the block feature matrix and setting a dynamic evolution interval to generate a topological feature matrix includes:

[0047] Perform regional weight distribution on the block feature matrix according to the material aging coefficient to generate a weighted feature matrix;

[0048] Perform dynamic threshold correction on the weighted feature matrix based on the ambient temperature and humidity data to generate a corrected feature matrix;

[0049] The modified characteristic matrix is ​​integrated in the time domain by the evolution rate function to generate the damage accumulation matrix;

[0050] The damage accumulation matrix is ​​compared point by point with a preset critical threshold to generate a dynamic evolution interval identification matrix.

[0051] Preferably, the step of inputting the repair objective function and the optimization constraints into a numerical iterative solver, calculating the repair parameters in a discrete space, and generating an initial sequence of repair parameters includes:

[0052] The finite difference method is used to discretize the repair objective function and generate a discrete optimization equation;

[0053] Construct the Jacobian matrix of the constraint conditions based on the dynamic equilibrium boundary and generate the constraint gradient information;

[0054] The discrete optimization equation is numerically solved by the Newton iteration algorithm to generate the initial values ​​of the repair parameters;

[0055] A boundary compliance check is performed on the initial values ​​of the repair parameters, parameters that exceed the constraint range are eliminated, and an initial sequence of repair parameters is generated.

[0056] Preferably, the process of converting the repair parameters into construction intensity values ​​through material property mapping to generate a new repair execution sequence includes:

[0057] Establish parameter-strength conversion function based on material elastic modulus and generate strength conversion coefficient matrix;

[0058] Perform element-by-element product operation on the restoration parameters and the intensity conversion coefficient matrix to generate a preliminary intensity value sequence;

[0059] The preliminary strength values ​​are limited based on the maximum load-bearing capacity of the construction equipment to generate a standardized strength series;

[0060] Performing time series smoothing filtering on the normalized intensity sequence to generate a new repair execution sequence.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] In terms of detection, comprehensive information on the status of underground structures can be obtained through the simultaneous multi-source acquisition of three-dimensional deformation data, crack distribution parameters, ambient temperature and humidity signals, and material stress response signals. This multi-source data acquisition approach avoids the limitations of a single detection method, making the detection results more accurate and reliable. After the collected data undergoes a series of operations such as spatial normalization, modal separation, and time-domain alignment correction, the generated structural damage characteristic tensor can accurately reflect the damage characteristics of the underground structure, effectively capturing even subtle damage changes and providing a solid data foundation for subsequent cracking status assessment.

[0063] In terms of cracking status assessment and repair decision-making, dynamic topological quantification of the structural damage characteristic tensor is performed, and the resulting damage evolution parameter matrix can clearly demonstrate the development trend of structural damage. Based on this, a multimodal coupling network is constructed, and by jointly training the repair decision network and the damage assessment network, it achieves real-time and accurate assessment of the cracking damage status of underground structures and generates targeted repair decision parameters. This process fully considers the interrelationships between multiple factors, making the assessment results more consistent with actual conditions and repair decisions more targeted, avoiding the blind repair problems of traditional methods and greatly improving repair efficiency and quality.

[0064] During the repair process, a structural damage constraint function was constructed based on the targeted repair decision parameters, and the dynamic equilibrium boundary of the cracked area was calculated. This boundary provided a clear scope and standard for the repair work, ensuring that the repair process was carried out under reasonable mechanical conditions and avoiding over- or under-repair. By iteratively optimizing the repair parameters and generating a targeted repair execution sequence, which was adjusted based on material properties and actual construction conditions, the repair work was better adapted to the specific needs of the underground structure, further improving the effectiveness and reliability of the repair.

[0065] The method also sets up a closed-loop control link. By comparing and analyzing the crack monitoring data with the target repair intensity, repair deviations are discovered in a timely manner and the repair parameters are dynamically corrected, thus achieving real-time monitoring and adjustment of the repair process. This closed-loop control mechanism can ensure that the repair work is always carried out in the direction of the expected goal, effectively improves the success rate of the repair, and ensures the long-term stability and safety of the underground structure. Overall, the present invention can significantly improve the accuracy and effectiveness of the diagnosis and treatment of cracks in underground structures in hot and humid areas, reduce maintenance costs, and ensure the safe operation of underground structures. It has important engineering application value and social benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a working principle diagram of the method for targeted diagnosis and treatment of underground structure cracks in hot and humid areas combined with non-destructive testing according to the present invention;

[0067] Figure 2 Flowchart for generating the damage evolution parameter matrix for dynamic topology quantification processing;

[0068] Figure 3 Flowchart for establishing a multimodal coupling network to evaluate cracking status and generate targeted repair decision parameters;

[0069] Figure 4 Flowchart for calculating the dynamic equilibrium boundary of the cracking area for constructing the structural damage constraint function;

[0070] Figure 5 Flowchart for process conversion and subsequent closed-loop control of a targeted repair execution sequence. DETAILED DESCRIPTION

[0071] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0072] See also Figure 1-Figure 5The present invention provides a targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing. The specific implementation steps are as follows:

[0073] Using professional non-destructive testing equipment, the three-dimensional deformation data, crack distribution parameters, environmental temperature and humidity signals, and material stress response signals of the underground structure are collected simultaneously. These data reflect the status of the underground structure from different angles and are the basis for subsequent analysis. The collected data form a non-destructive testing raw data matrix, which is then subjected to spatial normalization processing to eliminate the differences between the data due to different dimensions and value ranges for subsequent analysis. The standardized data is modally separated using a tensor decomposition algorithm to obtain the initial components of the damage characteristics. Since there may be slight differences in the data acquisition time, the initial components of the damage characteristics are aligned in the time domain, and the environmental coupling coefficient is introduced for phase compensation to generate more accurate damage characteristic components. Finally, based on the multi-scale fusion algorithm, the corrected damage characteristic components are correlated with the material degradation matrix for analysis, multi-dimensional features are extracted, and structural damage characteristic tensors under different working conditions are generated to comprehensively characterize the damage characteristics of the underground structure.

[0074] The structural damage feature tensor is quantified by regional blocks, and the damage distribution interval parameters are determined. Based on this, dynamic threshold calibration of the structural damage features is performed to generate a damage calibration matrix. The data in the structural damage feature tensor is then divided into multiple levels according to the damage calibration matrix to obtain a block feature matrix. Regional weights are assigned to the block feature matrix based on the material aging coefficient. The threshold is then dynamically adjusted using ambient temperature and humidity data. A topological feature matrix is ​​generated through time-domain integration of the evolution rate function and comparison with a preset critical threshold. The topological feature matrix is ​​subjected to environmental noise suppression and tensor normalization to obtain a damage evolution parameter matrix. A multimodal coupling network is constructed based on the damage evolution parameter matrix. The matrix is ​​divided into a repair feature matrix and an assessment feature matrix, respectively, to construct the repair decision network and the damage assessment network. The input layer of the repair decision network corresponds to deformation, crack, temperature and humidity, and stress signals, and the hidden layer uses a piecewise nonlinear activation function. The output layer generates the repair decision value. The input layer of the damage assessment network receives the output data of the repair network, and the hidden layer uses a hyperbolic tangent activation function. The output layer generates the damage quantification value. The two networks are jointly trained, and after residual function calculation and network robustness analysis, the network parameters are dynamically updated to finally generate targeted repair decision parameters.

[0075] An asymmetric matrix is ​​constructed based on the targeted repair decision parameters, and a damage state matrix is ​​obtained through linear transformation. Dynamic integral constraints are designed based on the damage state matrix, and equilibrium constraints are generated by performing a piecewise weighted operation on the damage variables. The two are combined to construct a structural stability function, which in turn yields a structural damage constraint function. The spatial gradient of the structural damage constraint function is calculated and compared with the modulus of the damage vector to generate damage constraints. Based on the damage constraints, the cracked area is partitioned into repair strength intervals, and a dynamic balance coefficient is set to determine the constraint range. The dynamic equilibrium boundary of the cracked area is obtained through boundary fitting, providing an important boundary basis for repair.

[0076] Based on the structural damage characteristic tensor, a repair weight matrix and a material compensation matrix are constructed and substituted into the dynamic optimization function to generate the repair objective function. Material strength and repair rate constraints are imposed on the repair objective function, and the dynamic equilibrium boundary is converted into a gradient convergence constraint to form an optimization constraint. The finite difference method is used to discretize the repair objective function, and the Jacobian matrix of the constraint condition is constructed to obtain gradient information. The initial values ​​of the repair parameters are solved using the Newton iterative algorithm, and the initial sequence of repair parameters is obtained after boundary compliance verification. The initial sequence is adaptively adjusted in step size, and stress constraints are processed in combination with the characteristics of the underground structure material. Finally, a targeted repair execution sequence is generated, providing specific operational steps for actual repair.

[0077] The technical solution of the present invention is further described in detail below through specific embodiments.

[0078] Example 1

[0079] In this embodiment, the process of collecting three-dimensional deformation data, crack distribution parameters, environmental temperature and humidity signals, and material stress response signals of underground structures and constructing a structural damage characteristic tensor is described in detail.

[0080] Utilizing specialized nondestructive testing equipment, we simultaneously collect 3D deformation data, crack distribution parameters, ambient temperature and humidity signals, and material stress response signals. These devices can include high-precision laser measuring instruments for acquiring 3D deformation data, crack observation instruments for measuring crack distribution parameters, temperature and humidity sensors for collecting ambient temperature and humidity signals, and stress sensors for collecting material stress response signals. Through simultaneous multi-source acquisition, we generate a matrix of raw nondestructive testing data.

[0081] The raw data matrix is ​​spatially normalized. This step aims to eliminate differences in dimensionality and numerical range between data sources, making the data comparable. This process yields a standardized data matrix, which is then subjected to modal separation using a tensor decomposition algorithm, decomposing the complex data into multiple initial components of damage signatures.

[0082] Perform time-domain alignment correction on the initial damage signature components. Because different types of data may have slight differences in acquisition time, phase compensation is performed by introducing an environmental coupling coefficient to ensure that the components accurately correspond in the time dimension, generating the corrected damage signature components.

[0083] Finally, a multi-scale fusion algorithm is used to correlate the corrected damage characteristic components with the material degradation matrix. The material degradation matrix reflects the degradation of material performance over time and environmental changes. This correlation analysis generates a damage correlation tensor, which is then subjected to multi-dimensional feature extraction to generate structural damage characteristic tensors under different operating conditions, providing a comprehensive and accurate data foundation for subsequent analysis and decision-making.

[0084] Consider an underground parking garage located in a hot and humid region. Due to long-term environmental impact, it has shown signs of structural cracking. A professional 3D laser scanner is used to scan the structure, collecting data at regular intervals (e.g., once an hour). The displacement changes at different locations in 3D space are captured, generating 3D deformation data. A high-precision crack observation instrument is used to carefully measure parameters such as the length, width, and spacing of cracks along the beams, columns, and walls of the parking garage, recording the crack distribution parameters. Multiple temperature and humidity sensors are installed in different areas of the parking garage to collect real-time ambient temperature and humidity signals, for example, recording temperature and humidity values ​​every 10 minutes. Simultaneously, stress sensors are placed at key stress-bearing locations to obtain material stress response signals and monitor stress changes within the structure. This data is collected synchronously to form a raw data matrix for nondestructive testing.

[0085] In the raw data matrix, the dimensions and numerical ranges of different data types vary significantly. For example, 3D deformation data is expressed in millimeters, crack width data is expressed in millimeters, temperature and humidity data are expressed in degrees Celsius and percentages, respectively, and stress data is expressed in megapascals. Spatial normalization is performed on the raw data, mapping all data types to a uniform interval (e.g., 0-1) to produce a standardized data matrix. Using a tensor decomposition algorithm, the standardized data matrix is ​​decomposed into multiple initial components of damage signatures, much like "breaking" complex structural damage data into more easily analyzable components.

[0086] Because different types of data may have slightly different collection times—for example, stress sensors collect data a few seconds later than temperature and humidity sensors—time-domain alignment correction is required for the initial damage signature components. This involves introducing an environmental coupling coefficient, which comprehensively accounts for the effects of temperature, humidity, and material properties on data acquisition time differences. Phase compensation is applied to each component to ensure accurate temporal alignment, generating the corrected damage signature components.

[0087] The material degradation matrix was constructed based on years of research data on the performance degradation of parking lot construction materials (such as concrete and steel) in hot and humid environments. Using a multi-scale fusion algorithm, the corrected damage characteristic components were correlated with the material degradation matrix and analyzed to generate a damage correlation tensor. Multidimensional features were extracted from this tensor, such as differences in damage severity across spatial regions and damage development trends over time. Ultimately, a structural damage characteristic tensor applicable to the underground parking lot under different operating conditions (e.g., normal use, after heavy rain, etc.) was generated.

[0088] Example 2

[0089] This embodiment describes in detail the process of performing dynamic topological quantization processing on the structural damage characteristic tensor, generating a damage evolution parameter matrix, establishing a multimodal coupling network to evaluate the cracking state, and generating targeted repair decision parameters.

[0090] First, the structural damage signature tensor is quantified by regional blocks. The entire structural damage signature tensor is divided into multiple regions according to specific rules, and the damage distribution interval parameters of each region are calculated. Based on these parameters, the structural damage signature is dynamically thresholded, and the threshold range corresponding to different damage levels is determined to generate a damage calibration matrix.

[0091] Subsequently, the damage feature data in the structural damage feature tensor is divided into multiple levels according to the damage calibration matrix. According to the different levels of damage degree, the data is classified to generate a block feature matrix.

[0092] Next, regional weights are assigned to the block feature matrix based on the material aging coefficient. The material aging coefficient reflects the degradation of material properties over time, with higher weights assigned to regions with higher aging levels. After generating the weighted feature matrix, a dynamic threshold correction is applied based on ambient temperature and humidity data. Temperature and humidity fluctuations can affect the progression of structural damage. The threshold is adjusted based on real-time temperature and humidity data to generate a corrected feature matrix.

[0093] The modified characteristic matrix is ​​then integrated in the time domain using the evolution rate function to calculate the damage accumulation in the time dimension and generate a damage accumulation matrix. The damage accumulation matrix is ​​then compared point by point with the preset critical threshold to determine whether each point reaches or exceeds the threshold. This generates a dynamic evolution interval identification matrix, which in turn generates a topological characteristic matrix.

[0094] The topological feature matrix is ​​passed through a nonlinear filter to suppress environmental noise, remove noise interference from the data, and generate a noise-reduced feature matrix. The noise-reduced feature matrix is ​​then subjected to tensor normalization transformation to meet the requirements of subsequent analysis and generate a damage evolution parameter matrix.

[0095] Based on the damage evolution parameter matrix, a multimodal coupling network is constructed. The damage evolution parameter matrix is ​​divided into a repair feature matrix and an assessment feature matrix, and a repair decision network and a damage assessment network are constructed, respectively. The input layer of the repair decision network contains four neurons corresponding to deformation, crack, temperature and humidity, and stress signals. The hidden layer uses a piecewise nonlinear activation function, and the output layer generates the repair decision value, which is the output data of the repair network. The input layer of the damage assessment network receives the output data of the repair network. The hidden layer uses a hyperbolic tangent activation function, and the output layer generates the damage quantification value, which is the output data of the assessment network.

[0096] The repair decision network and damage assessment network are jointly trained to optimize network parameters and generate a multimodal coupled network structure. Residual function calculation and network robustness analysis are performed on the multimodal coupled network structure to evaluate network performance. Based on the evaluation results, the parameters of the repair decision network and damage assessment network are dynamically updated. The network weights are adjusted using a backpropagation algorithm to ultimately generate targeted repair decision parameters.

[0097] For example, a subway tunnel located in a humid and hot region has experienced varying degrees of structural damage over its long-term operation. First, the structural damage characteristic tensor of the subway tunnel is divided into multiple regions of a certain length (e.g., 10 meters). The damage distribution interval parameters within each region are calculated, such as the maximum and minimum crack widths and frequency within that region. Based on these parameters, thresholds corresponding to different damage levels are determined. For example, a crack width greater than 1 mm is considered severe damage, 0.5-1 mm is considered moderate damage, and less than 0.5 mm is considered mild damage. This is used to generate a damage calibration matrix.

[0098] Based on the damage calibration matrix, the data in the tunnel structure damage feature tensor is divided into multiple levels. For example, data such as crack width and deformation are classified according to the standards of mild, moderate, and severe damage, forming data blocks of different levels, and then generating a block feature matrix.

[0099] The material aging coefficient is determined based on the service life of subway tunnels and material aging research data. Areas with longer service life have higher material aging coefficients. Regional weights are assigned to the block feature matrix, with higher weights assigned to areas with higher aging coefficients. Dynamic threshold corrections are then performed on the weighted feature matrix using real-time temperature and humidity data within the tunnel. If the humidity within the tunnel suddenly increases, the crack width damage threshold is appropriately lowered, as increased humidity can accelerate crack development. This results in a revised feature matrix.

[0100] The modified feature matrix is ​​integrated over time using a predefined evolution rate function. For example, the daily damage accumulation is calculated over a one-day interval to generate a damage accumulation matrix. This matrix is ​​then compared point by point with a preset critical threshold (e.g., if the damage accumulation in a region exceeds a certain value, it warrants attention). This determines whether each region has reached or exceeded the threshold, generating a dynamic evolution interval identification matrix and ultimately, a topological feature matrix.

[0101] The topological feature matrix is ​​processed using a nonlinear filter to remove environmental noise caused by sensor errors and other factors, generating a noise-reduced feature matrix. The noise-reduced feature matrix is ​​then subjected to tensor normalization transformation to ensure that the data meets the requirements of subsequent analysis, resulting in the damage evolution parameter matrix.

[0102] The damage evolution parameter matrix is ​​divided into a repair feature matrix and an assessment feature matrix. A repair decision network is constructed based on the repair feature matrix. The four neurons in the input layer correspond to tunnel deformation, cracks, temperature and humidity, and stress signals, respectively. The hidden layer uses a piecewise nonlinear activation function, such as the ReLU function, to enable the network to better learn data features. The output layer generates repair decision values, such as the recommended repair method (local repair or overall reinforcement), which are then output data from the repair network. A damage assessment network is constructed based on the assessment feature matrix. The input layer receives the output data from the repair network, and the hidden layer uses the hyperbolic tangent activation function. The output layer generates a damage quantification value reflecting the severity of the current tunnel damage, which is then output data from the assessment network.

[0103] The repair decision network and damage assessment network are jointly trained, with network parameters continuously adjusted to generate a multimodal coupled network structure. The network robustness is analyzed by calculating the difference between the network's predicted and actual values ​​using a residual function. Based on the analysis results, the network weights are dynamically updated using a backpropagation algorithm, ultimately generating targeted repair decision parameters for the subway tunnel, such as determining key repair areas and the amount of repair material required.

[0104] Example 3

[0105] This embodiment focuses on the implementation process of constructing a structural damage constraint function based on targeted repair decision parameters and calculating the dynamic equilibrium boundary of the cracked area.

[0106] An asymmetric matrix is ​​constructed for the targeted repair decision parameters. Based on the characteristics and relationships of the decision parameters, a damage state matrix is ​​generated through a specific linear transformation. This matrix can intuitively reflect the current damage state of the underground structure.

[0107] Dynamic integral constraints are designed based on the damage state matrix. A piecewise weighting operation is performed on the damage variables, assigning different weights according to the degree of damage to generate equilibrium constraints. The damage state matrix and equilibrium constraints are combined to construct a structural stability function, thereby generating a structural damage constraint function that comprehensively considers the damage state and stability requirements of the structure.

[0108] The spatial gradient of the structural damage characteristic tensor is calculated to obtain the gradient analysis results. The gradient analysis results are compared with the modulus of the damage vector to determine the spatial variation of the damage and generate damage constraints.

[0109] Based on the damage constraints, the cracked area is partitioned into repair strength intervals. Dynamic equilibrium coefficients are set based on the actual structure to determine the constraint range. Based on the constraint range, boundary fitting is performed using appropriate mathematical methods, such as curve fitting algorithms, to generate the dynamic equilibrium boundary of the cracked area. This boundary defines the required strength range and boundary conditions for repair under different damage conditions, providing an important basis for subsequent repair work.

[0110] Consider an underground warehouse built many years ago in a hot and humid region. Its structure is cracking, and the dynamic equilibrium boundary of the cracked area needs to be determined. The acquired targeted repair decision parameters, such as the required strength of the repair material and the required repair time, are used to construct an asymmetric matrix through a specific linear transformation to form a damage state matrix. This can be expressed as: ,in represents the damage state matrix, is the targeted repair decision parameter, Represents a specific linear transformation operation, which is determined according to the relationship between the structural characteristics of the underground warehouse and the repair decision parameters.

[0111] Based on the damage state matrix , perform piecewise weighting operations on damage variables (such as crack width, structural deformation, etc.). For example, a lower weight is assigned to areas with crack widths less than 0.5 mm, and a higher weight is assigned to areas with crack widths greater than 0.5 mm, generating equilibrium constraints. The structural stability function is constructed by combining the equilibrium constraint term, that is, the structural damage constraint function .

[0112] Calculating structural damage constraint functions The spatial gradient of the crack is calculated to obtain the gradient analysis result. This result is compared with the modulus of the damage vector (a vector composed of crack width, deformation, etc.) to determine the spatial variation of the damage and obtain the damage constraint condition.

[0113] Based on damage constraints, the cracked area of ​​the underground warehouse is divided into repair strength zones. For example, they are divided into low-strength, medium-strength, and high-strength repair zones. A dynamic balance coefficient is set, taking into account factors such as the warehouse's operational requirements and structural load-bearing capacity. Based on the zones and the dynamic balance coefficient, a constraint range is determined. A curve fitting algorithm is used to fit the constraint range's boundaries, ultimately generating a dynamic balance boundary for the cracked area, providing a key basis for subsequent repairs.

[0114] Example 4

[0115] This embodiment focuses on describing the specific operations of iteratively optimizing repair parameters based on the structural damage characteristic tensor and the dynamic equilibrium boundary to generate a targeted repair execution sequence.

[0116] Based on the structural damage characteristic tensor, a repair weight matrix and a material compensation matrix are constructed. The repair weight matrix reflects the importance of different damage locations to the repair work, while the material compensation matrix accounts for differences in material properties. Substituting the repair weight matrix and the material compensation matrix into the dynamic optimization function generates the repair objective function, which aims to achieve the optimal repair of the underground structure.

[0117] Material strength constraints and repair rate constraints are imposed on the repair objective function. The material strength constraint ensures that the material strength used during the repair process meets structural requirements, while the repair rate constraint prevents the repair work from being too hasty or too slow, which could affect the repair effect and project progress. Furthermore, the dynamic equilibrium boundary is converted into a gradient convergence constraint to generate optimization constraints, further limiting the range of values ​​for the repair parameters.

[0118] The repair objective function and optimization constraints are input into a numerical iterative solver. The repair objective function is discretized using the finite difference method, converting the continuous function into a discrete equation to generate a discrete optimization equation. The Jacobian matrix of the constraints is constructed based on the dynamic equilibrium boundary, and the constraint gradient information is obtained. The discrete optimization equation is numerically solved using the Newton iterative algorithm to obtain the initial values ​​of the repair parameters. The initial values ​​of the repair parameters are then checked for boundary compliance, and parameters that exceed the constraint range are eliminated to generate an initial sequence of repair parameters.

[0119] Adaptive step-size adjustments are performed on the initial sequence of repair parameters, dynamically adjusting the iteration step size based on actual conditions to improve computational efficiency and accuracy. Stress constraints are applied based on the material properties of underground structures to ensure that the stress state of the structure remains within a safe range during the repair process. Ultimately, a targeted repair execution sequence is generated, providing specific operational steps and parameters for the actual repair work.

[0120] For example, consider an underground sewage treatment tank located in a hot and humid region. Cracks have developed in the tank and require repair. Based on the tank's structural damage characteristic tensor, a repair weight matrix and a material compensation matrix are constructed. For example, areas near the sewage inlet and outlet are more severely damaged due to erosion and corrosion, and are therefore given a higher weight in the repair weight matrix. For the special corrosion-resistant materials used in the tank, a material compensation matrix is ​​constructed based on their properties. These two matrices are then substituted into a dynamic optimization function to generate a repair objective function, the goal of which is to achieve optimal performance after the repair.

[0121] Material strength constraints and repair rate constraints are imposed on the repair objective function. The material strength constraint requires that the strength of the repair material used be no less than that of the original pool material. The repair rate constraint limits the area repaired per day to a certain value, preventing repairs that are too fast and affect quality, or too slow and delay the project. Furthermore, the dynamic equilibrium boundary is converted into a gradient convergence constraint to form an optimization constraint.

[0122] The repair objective function and optimization constraints are input into a numerical iterative solver. The repair objective function is discretized using the finite difference method, converting the continuous function into discrete points for calculation, resulting in a discrete optimization equation. The Jacobian matrix of the constraints is constructed based on the dynamic equilibrium boundary, and constraint gradient information is obtained. The discrete optimization equation is numerically solved using the Newton iteration algorithm to obtain initial values ​​for the repair parameters. These initial values ​​are then checked for boundary compliance, and parameters outside the constraint range, such as those outside the material strength range or repair rate range, are eliminated to generate an initial sequence of repair parameters.

[0123] Adaptively adjust the step size of the initial sequence of repair parameters. In the early stages of iteration, the step size can be set larger to speed up the calculation; as the optimal solution is approached, the step size is reduced to improve accuracy. Stress constraints are applied based on the material properties of the sewage treatment tank to ensure that the tank stress remains within a safe range during the repair process. Ultimately, a targeted repair execution sequence is generated, clarifying the specific repair steps and parameters, such as the amount of repair material used in each area and the repair sequence.

[0124] Example 5

[0125] This embodiment describes in detail the subsequent process of completing the targeted repair of underground structure cracks after generating the targeted repair execution sequence.

[0126] The targeted repair execution sequence is converted into a process, and the repair parameters are converted into specific construction operation instructions, such as construction sequence and construction method. Error compensation is performed on the construction control instructions, taking into account various error factors that may occur during the actual construction process, and compensation is performed through a certain algorithm to generate a repair execution signal.

[0127] The target repair strength is calculated based on the stability requirements of the underground structure. This strength is a key indicator for ensuring the stable operation of the structure after repair. At the same time, the crack growth rate is monitored in real time, and real-time data on crack growth is obtained through professional monitoring equipment to generate crack monitoring data.

[0128] The difference between the crack monitoring data and the target repair intensity is calculated to obtain a repair deviation value. This repair deviation value is then compared with a preset repair threshold to generate a deviation determination result. If the deviation value exceeds the threshold, it indicates that there is a problem with the repair work and adjustments are required.

[0129] Based on the deviation determination results, the repair parameters are dynamically corrected. If the deviation exceeds the range, the repair parameters are adjusted using an appropriate algorithm based on the size and direction of the deviation to generate updated repair parameters. The updated repair parameters are converted into a construction sequence, and a parameter-strength conversion function is established based on the material elastic modulus to generate a strength conversion coefficient matrix. The repair parameters are element-by-element multiplied with the strength conversion coefficient matrix to generate a preliminary strength value sequence. The preliminary strength values ​​are limited based on the maximum load-bearing capacity of the construction equipment to ensure that the construction strength is within the equipment's tolerance range, generating a standardized strength sequence. The standardized strength sequence is then subjected to time series smoothing filtering to remove fluctuations in the data and generate a new repair execution sequence.

[0130] The new repair execution sequence is input into the repair execution device for closed-loop control, dynamically adjusting the construction process based on real-time monitoring data. This process is repeated until the repair deviation value falls within the preset threshold, completing the targeted repair of the underground structure cracks and ensuring its safety and stability.

[0131] For example, an underground utility corridor in a hot and humid region was found to have cracks in some of its structures during routine inspections, and repairs are now underway. The generated targeted repair execution sequence undergoes process conversion. For example, if the repair execution sequence includes a section of the utility corridor requiring local repair, the process conversion converts it into specific construction operation instructions, including the type of repair material to use, the repair process to be adopted (such as grouting or painting), and the construction personnel's operating procedures. Because errors may exist during the actual construction process, error compensation is performed on the construction control instructions. For example, considering the precision errors of construction equipment and the inconsistencies in construction personnel's operations, the instructions are adjusted using a specific algorithm to generate a repair execution signal.

[0132] The target repair strength is calculated based on the design load-bearing capacity and stability requirements of the underground utility corridor. Assuming the corridor design requires that structural deformation cannot exceed a certain value when subjected to a certain pressure, the desired strength of the repaired structure is calculated based on this requirement. Simultaneously, sensors monitor the crack growth rate in real time, recording the crack extension length at regular intervals (e.g., half an hour) to generate crack monitoring data.

[0133] The difference between the crack monitoring data and the target repair intensity is calculated. For example, if the crack width corresponding to the target repair intensity should be stable within 0.2 mm, and the current crack width is 0.3 mm, the difference between the two is the repair deviation value. The repair deviation value is compared with the preset repair threshold (such as 0.1 mm). If the deviation value exceeds the threshold, it indicates that there is a problem with the repair work and adjustments are needed.

[0134] The repair parameters are dynamically corrected based on the deviation judgment results. If the deviation exceeds the range, the repair parameters are adjusted using an appropriate algorithm according to the size and direction of the deviation. For example, if the crack width is too large, the amount of repair material should be appropriately increased. The updated repair parameters are converted into a construction sequence, and a parameter-strength conversion function is established based on the elastic modulus of the tunnel construction material to generate a strength conversion coefficient matrix. The repair parameters are multiplied element by element with the strength conversion coefficient matrix to obtain a preliminary strength value sequence. Taking into account the maximum load-bearing capacity of the construction equipment, the preliminary strength value is limited. For example, the maximum strength that the construction equipment can apply is a certain value, and the excess is adjusted to generate a standardized strength sequence. The standardized strength sequence is time-series smoothed and filtered to remove data fluctuations and generate a new repair execution sequence.

[0135] The new repair execution sequence is input into a repair execution device, such as automated repair equipment. Based on real-time monitoring data, the construction process is dynamically adjusted. If the repair effect on a particular tunnel section is found to be poor, the injection rate of the repair material or the operating frequency of the repair equipment is promptly adjusted. This process is repeated until the repair deviation value falls within the preset threshold, completing the targeted repair of the underground tunnel crack and ensuring its safe and stable operation.

[0136] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0137] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing, characterized in that: The following steps are involved: Collect 3D deformation data, crack distribution parameters, ambient temperature and humidity signals, and material stress response signals of underground structures, and construct a structural damage characteristic tensor; Performing dynamic topological quantization on the structural damage characteristic tensor to generate a damage evolution parameter matrix, and establishing a multimodal coupling network to evaluate the cracking state and generate targeted repair decision parameters; Constructing a structural damage constraint function according to the targeted repair decision parameters, and calculating a dynamic equilibrium boundary of the cracked area; Iteratively optimize repair parameters based on the structural damage characteristic tensor and the dynamic equilibrium boundary to generate a targeted repair execution sequence; The dynamic topological quantization processing of the structural damage characteristic tensor is performed to generate a damage evolution parameter matrix, and a multimodal coupling network is established to evaluate the cracking state and generate targeted repair decision parameters, including: Performing regional block quantization on the structural damage feature tensor to generate damage distribution interval parameters, and performing dynamic threshold calibration on the structural damage feature based on the damage distribution interval parameters to generate a damage calibration matrix; Performing multi-level division processing on the damage feature data in the structural damage feature tensor according to the damage calibration matrix to generate a block feature matrix; Performing weighted topological processing on the block feature matrix and setting a dynamic evolution interval to generate a topological feature matrix; The topological characteristic matrix is ​​subjected to environmental noise suppression by a nonlinear filter to generate a noise reduction characteristic matrix, and the noise reduction characteristic matrix is ​​subjected to tensor normalization conversion to generate a damage evolution parameter matrix; Based on the damage evolution parameter matrix, a multimodal coupling network is constructed, and the cracking damage state of the underground structure is evaluated in real time to generate targeted repair decision parameters.

2. The targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing according to claim 1 is characterized in that: The collecting of three-dimensional deformation data, crack distribution parameters, environmental temperature and humidity signals, and material stress response signals of the underground structure, and constructing a structural damage characteristic tensor, includes: Perform multi-source synchronous acquisition of 3D deformation data, crack distribution parameters, ambient temperature and humidity signals, and material stress response signals to generate a non-destructive testing raw data matrix; Performing spatial normalization processing on the original data matrix to obtain a standardized data matrix, and performing modal separation on the standardized data matrix based on a tensor decomposition algorithm to generate initial components of damage characteristics; Performing time domain alignment correction on the initial damage characteristic component, performing phase compensation by introducing an environmental coupling coefficient, and generating a corrected damage characteristic component; Based on a multi-scale fusion algorithm, the corrected damage characteristic components and the material degradation matrix are correlated and analyzed to generate a damage correlation tensor. Multi-dimensional features are extracted from the damage correlation tensor to generate structural damage characteristic tensors under different working conditions.

3. The targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing according to claim 1 is characterized in that: The multimodal coupling network is constructed based on the damage evolution parameter matrix, and the crack damage state of the underground structure is evaluated in real time to generate targeted repair decision parameters, including: Dividing the damage evolution parameter matrix into a repair feature matrix and an evaluation feature matrix; A repair decision network is constructed based on the repair feature matrix. The repair decision network includes an input layer, a hidden layer, and an output layer. The input layer contains four neurons corresponding to deformation, crack, temperature and humidity, and stress signals. The hidden layer uses a piecewise nonlinear activation function. The output layer generates a repair decision value to obtain the repair network output data. Constructing a damage assessment network based on the assessment feature matrix, the damage assessment network includes an input layer, a hidden layer, and an output layer, wherein the input layer receives the output data of the repair network, the hidden layer uses a hyperbolic tangent activation function, and the output layer generates a damage quantization value to obtain the output data of the assessment network; Jointly training the repair decision network and the damage assessment network to generate a multimodal coupled network structure, and performing residual function calculation and network robustness analysis on the multimodal coupled network structure to generate a network evaluation result; Based on the network evaluation results, the parameters of the repair decision network and the damage assessment network in the multimodal coupling network structure are dynamically updated, and the network weights are adjusted through the back propagation algorithm to generate targeted repair decision parameters.

4. The targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing according to claim 1 is characterized in that: The step of constructing a structural damage constraint function based on the targeted repair decision parameters and calculating the dynamic equilibrium boundary of the cracked area includes: Performing an asymmetric matrix construction on the targeted repair decision parameters and generating a damage state matrix through linear transformation; Designing dynamic integral constraint terms based on the damage state matrix, performing piecewise weighting operations on damage variables to generate equilibrium constraint terms, and combining the damage state matrix and the equilibrium constraint terms to construct a structural stability function to generate a structural damage constraint function; Calculating the spatial gradient of the structural damage constraint function to generate a gradient analysis result, and performing a comparison operation on the gradient analysis result and the modulus length of the damage vector to generate a damage constraint condition; The repair strength interval of the cracked area is partitioned according to the damage constraint condition, and a dynamic balance coefficient is set to generate a constraint range. Boundary fitting is performed based on the constraint range to generate a dynamic balance boundary of the cracked area.

5. The targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing according to claim 1 is characterized in that: The iterative optimization of repair parameters based on the structural damage characteristic tensor and the dynamic equilibrium boundary to generate a targeted repair execution sequence includes: Constructing a repair weight matrix and a material compensation matrix according to the structural damage characteristic tensor, and substituting the repair weight matrix and the material compensation matrix into a dynamic optimization function to generate a repair objective function; Applying material strength constraints and repair rate constraints to the repair objective function, and converting the dynamic equilibrium boundary into a gradient convergence constraint to generate an optimization constraint; Inputting the repair objective function and the optimization constraints into a numerical iterative solver, calculating the repair parameters in a discrete space, and generating an initial sequence of repair parameters; Adaptive step size adjustment is performed on the initial sequence of repair parameters, and stress constraint processing is performed in combination with the material properties of the underground structure to generate a targeted repair execution sequence.

6. The targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing according to claim 1 is characterized in that: The method further comprises: Performing process conversion on the targeted repair execution sequence to generate a construction control instruction, and performing error compensation on the construction control instruction to generate a repair execution signal; Calculate the target repair strength based on the stability requirements of the underground structure, monitor the crack expansion rate in real time, and generate crack monitoring data; Calculating the difference between the crack monitoring data and the target repair strength value to generate a repair deviation value, and comparing the repair deviation value with a preset repair threshold to generate a deviation determination result; Dynamically correcting the repair parameters based on the deviation determination result to generate updated repair parameters, and performing construction sequence conversion on the updated repair parameters, converting the repair parameters into construction intensity values ​​through material property mapping to generate a new repair execution sequence; The new repair execution sequence is input into the repair execution device for closed-loop control, and the construction process is dynamically adjusted to complete the targeted repair of underground structure cracks.

7. The targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing according to claim 1 is characterized in that: The weighted topological processing is performed on the block feature matrix, and a dynamic evolution interval is set to generate a topological feature matrix, including: Perform regional weight distribution on the block feature matrix according to the material aging coefficient to generate a weighted feature matrix; Perform dynamic threshold correction on the weighted feature matrix based on the ambient temperature and humidity data to generate a corrected feature matrix; The modified characteristic matrix is ​​integrated in the time domain by the evolution rate function to generate the damage accumulation matrix; The damage accumulation matrix is ​​compared point by point with a preset critical threshold to generate a dynamic evolution interval identification matrix.

8. The targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing according to claim 5, characterized in that: The step of inputting the restoration objective function and the optimization constraint conditions into a numerical iterative solver, calculating the restoration parameters in a discrete space, and generating an initial sequence of restoration parameters includes: The finite difference method is used to discretize the repair objective function and generate a discrete optimization equation; Construct the Jacobian matrix of the constraint conditions based on the dynamic equilibrium boundary and generate the constraint gradient information; The discrete optimization equation is numerically solved by the Newton iteration algorithm to generate the initial values ​​of the repair parameters; A boundary compliance check is performed on the initial values ​​of the repair parameters, parameters that exceed the constraint range are eliminated, and an initial sequence of repair parameters is generated.

9. The targeted diagnosis and treatment method for underground structure cracks in hot and humid areas combined with non-destructive testing according to claim 6, characterized in that: The repair parameters are converted into construction intensity values ​​through material property mapping to generate a new repair execution sequence, including: Establish parameter-strength conversion function based on material elastic modulus and generate strength conversion coefficient matrix; Perform element-by-element product operation on the restoration parameters and the intensity conversion coefficient matrix to generate a preliminary intensity value sequence; The preliminary strength values ​​are limited based on the maximum load-bearing capacity of the construction equipment to generate a standardized strength series; Performing time series smoothing filtering on the normalized intensity sequence to generate a new repair execution sequence.

Images