Multi-environment coupling degradation prediction and adaptive repair collaborative optimization method and system

Abstract

The invention discloses a multi-environment coupling degradation prediction and adaptive repair collaborative optimization method and system. A first sensor network is arranged on the surface of a structure to collect environment and material performance data, a degradation prediction model is constructed, a quantum annealing algorithm and an artificial neural network are used for global optimization, and a high-precision degradation prediction model is obtained; meanwhile, a second sensor network and a self-adaptive repairing material are integrally embedded in the structure, crack, strain and corrosion data are collected in real time, when a monitoring index exceeds a threshold value, microcapsules are triggered to be broken to achieve self-repairing, and strength data after repairing are returned to the cloud end. And the cloud end adopts a multi-objective optimization algorithm to iteratively generate an optimal repair strategy based on the repair efficiency, the cost and the number of damage points, and synchronously performs incremental updating on the high-precision degradation prediction model to form a'perception-prediction-repair-feedback 'closed loop until the health index and the residual life reach the standard. The full-life-cycle intelligent and automatic repair of the material can be realized.

Description

Technical Field

[0001] This invention relates to the field of materials science and engineering, specifically to a method and system for synergistic optimization of multi-environment coupled degradation prediction and adaptive repair. Background Technology

[0002] In modern engineering, the prediction and repair of material degradation are key factors in ensuring the long-term safety and reliability of engineering structures. Currently, many existing technologies for material degradation prediction and repair strategies suffer from the following problems:

[0003] (1) The accuracy of degradation prediction is low.

[0004] Existing technologies typically rely on modeling the impact of single environmental factors (such as temperature and humidity) on material degradation, neglecting the complex coupling effects between environmental factors. For example, traditional degradation prediction methods often only consider the impact of temperature changes on materials, failing to fully account for the synergistic effects of multiple factors such as temperature, humidity, salinity, and pressure. This results in insufficient prediction accuracy of existing methods in complex environments (such as the deep sea, polar regions, and outer space), failing to meet the application requirements of demanding engineering projects.

[0005] (2) The repair strategy lacks intelligence and real-time capability.

[0006] Current repair methods typically rely on human experience and static data, lacking intelligent and real-time feedback mechanisms in the repair decision-making process. Many existing repair systems fail to monitor the health status of materials in real time and lack the ability to automatically adjust repair strategies based on real-time data. This can lead to delayed or over-repair measures, increasing costs and affecting repair effectiveness.

[0007] (3) Limited application scope and environmental adaptability

[0008] Most existing material health management and repair technologies are only applicable to conventional environments (such as normal climatic conditions) and cannot effectively cope with extreme environments (such as deep-sea, high-radiation, high-temperature, or extremely low-temperature environments). For example, the high-pressure and corrosive environments in deep-sea engineering, the extremely low-temperature and high-humidity environments in polar engineering, and the radiation and extreme temperature differences in space all pose significant challenges to material degradation and repair. Existing technologies have limited adaptability and scalability, and cannot meet the practical needs of these complex environments.

[0009] (4) Lack of comprehensive life cycle management

[0010] Currently, many material health management systems lack comprehensive management of the entire material lifecycle, especially in the long-term effectiveness evaluation and repair scheme optimization after repair. Existing technologies generally do not achieve dynamic adjustment. The lack of closed-loop control between material health monitoring and repair effect feedback leads to significant uncertainties in the repair process, making it impossible to achieve long-term stability and continuous optimization.

[0011] These issues limit the application of existing technologies in extreme environments and fail to provide efficient, intelligent, and adaptable material health management and repair solutions. Therefore, there is an urgent need for a new technological solution that can combine multiple environmental factors to provide accurate degradation prediction, intelligent repair strategies, and comprehensive lifecycle management to meet the material health management needs of modern engineering, especially in complex environments.

[0012] In summary, through analysis of existing technologies, it can be seen that the technical solution of this invention has significant advantages in terms of degradation prediction accuracy, intelligent repair strategies, environmental adaptability, and lifecycle management. Compared with existing technologies, this invention can provide more efficient, accurate, and intelligent solutions in complex environments, overcoming the limitations of existing technologies. Summary of the Invention

[0013] To address the problems existing in the prior art, this invention provides a multi-environment coupled degradation prediction and adaptive repair collaborative optimization method and system. The purpose is to significantly improve the accuracy of degradation prediction by adopting a degradation prediction model coupled with multiple environmental loads, comprehensively considering the influence of multiple factors such as temperature, humidity, salinity, and pressure on the performance of the target structural material, and combining quantum computing and artificial intelligence optimization.

[0014] To achieve the above objectives, the specific solution of the present invention is as follows:

[0015] A multi-environment coupled degradation prediction and adaptive repair co-optimization method includes the following steps:

[0016] Step 1: Collect environmental factor data and material performance data by deploying a first sensor network on the surface of the target structure, and construct a degradation prediction model that includes the degradation coefficient and degradation index to be optimized for materials coupled with multiple environmental loads; the environmental factor data includes temperature, humidity, salinity and pressure, and the material performance data includes initial compressive strength, elastic modulus, coefficient of thermal expansion and porosity.

[0017] Step 2: Using the environmental factor data and material performance data collected in Step 1 as the original samples, and combining them with the spatiotemporal degradation history data corresponding to the original samples, a standardized training dataset is constructed after handling missing values ​​and standardization. The degradation coefficient and degradation index to be optimized in the degradation prediction model of Step 1 are globally optimized using the quantum annealing algorithm. Then, the optimized degradation coefficient and degradation index are fixed, and an artificial neural network is used to establish a nonlinear mapping of environmental performance to obtain a high-precision degradation prediction model. The degradation history data includes the degradation rate, crack propagation, and strength loss of materials under different environments.

[0018] Step 3: Embed an adaptive repair material and deploy a second sensor network in the target structure. The adaptive repair material includes a self-healing polymer matrix, a repair agent, and a reinforcing material. The self-healing polymer matrix is ​​epoxy resin or polyurethane resin, the repair agent is embedded in the self-healing polymer matrix in the form of microcapsules or fiber mesh, and the reinforcing material is carbon nanotubes or graphene. The second sensor network is integrated with the adaptive repair material and real-time data on crack width, strain value, and corrosion rate are collected as health data. When the second sensor network detects cracks or strength loss exceeding a threshold, the microcapsules rupture and release the repair agent. Simultaneously, post-repair strength recovery data is continuously collected and transmitted back to the cloud big data platform. The repair efficiency is calculated based on the post-repair strength recovery data, and an initial repair strategy is generated using an optimization objective function, taking into account repair cost, strength change, strain change, and number of damage points.

[0019] Step 4: Write the high-precision degradation prediction model from Step 2 and the repair efficiency from Step 3 into the simulation platform. Input real-time environmental factor data, material performance data, and historical degradation data accumulated from Step 3 to predict the material life and performance degradation trend. Based on the initial repair strategy generated in Step 3, evaluate the effect of each repair scheme on material strength recovery and life extension to obtain life prediction results and repair effect evaluation results. Use optimization algorithms to perform multi-objective optimization on repair time and repair material type to form the optimal repair strategy.

[0020] Step 5: Perform repair on the target structure according to the optimal repair strategy obtained in Step 4, and feed the repair effect back to the high-precision degradation prediction model in Step 2 for dynamic updating of model parameters. With repair cost, strength change, strain change, and number of damage points as optimization objectives, use the same multi-objective optimization algorithm as in Step 4 to iteratively optimize the repair time, repair materials, and repair methods to obtain a dynamically updated repair strategy. Calculate the health index based on the material compressive strength at time t and the initial compressive strength output by the high-precision degradation prediction model in Step 2, and calculate the remaining life based on the health index and the degradation coefficient obtained in Step 2.

[0021] Step 6, closed-loop optimization and repair feedback update based on big data, includes the following sub-steps:

[0022] Step 6.1: Upload the high-precision degradation prediction model obtained in Step 2, the repair efficiency obtained in Step 3, and the optimal repair strategy obtained in Step 4 to the cloud big data platform.

[0023] Step 6.2: Continuously acquire environmental factor data, material performance data, crack width, strain value, corrosion rate, and post-repair strength recovery data through the first sensor network in Step 1 and the second sensor network in Step 3. After data cleaning, missing value filling, and normalization, new samples are formed.

[0024] Step 6.3: Use the new samples from Step 6.2 to incrementally update the degradation coefficient and degradation index in Step 2, and then send the updated degradation coefficient and degradation index back to the high-precision degradation prediction model in Step 2 to replace the original fixed values ​​and dynamically refresh the model parameters.

[0025] Step 6.4: Input the high-precision degradation prediction model refreshed in Step 6.3 back into the simulation platform in Step 4. Combine it with the strength recovery data accumulated in Step 6.2 after repair, recalculate the repair efficiency, and use the same multi-objective optimization algorithm as in Step 4 to iteratively optimize the repair time and the type of repair material. The optimization result is used as the new current optimal repair strategy.

[0026] Step 6.5: After performing the repair according to the new optimal repair strategy in Step 6.4, continue to collect the strength recovery data, strain value and crack width after repair through the second sensor network. Repeat steps 6.2 to 6.4 with the newly added data to obtain the updated current optimal repair strategy.

[0027] Step 6.6: Repeat steps 6.2 to 6.5 to make the high-precision degradation prediction model in step 2 converge continuously with the measured data until the health index output in step 2 is greater than or equal to the preset threshold and the remaining lifespan is greater than or equal to the preset lifespan value.

[0028] Furthermore, the formula for constructing the environmental load-material coupled degradation model described in step 1 is as follows:

[0029] (1),

[0030] In the formula: express The compressive strength of the material at all times; Indicates the initial compressive strength; This function represents the influence of environmental factors on material properties, specifically the effects of temperature, humidity, salinity, and pressure. It is fitted to environmental data to describe the degradation effects of these factors on the material. The degradation coefficient, expressed in units of 1 / year, describes the rate of material degradation under harsh environments and depends on environmental factors and the type of material; the degradation rate under different environments is obtained through experiments. Indicates time, in years, indicating the service life of the material; The degradation index represents the degree of acceleration of material degradation; this degradation index is obtained by fitting experimental data under different material and environmental conditions. The values ​​are different.

[0031] Furthermore, the formula for the quantum degeneration algorithm described in step 2 is as follows:

[0032] (4),

[0033] In the formula: Corresponding energy state The probability of; It represents the energy state, specifically the optimization target of the degradation coefficient and degradation index; Represents the Boltzmann constant; This represents the temperature change during quantum annealing; represents the partition function, used to normalize the probability distribution; e represents the base of the natural logarithm, which is 2.71828, used for exponential calculation;

[0034] The artificial neural network is a deep neural network or a convolutional neural network. The deep neural network is used to perform regression prediction on the compressive strength and degradation rate of materials, using the mean squared error as the loss function. The formula for the mean squared error is as follows:

[0035] (5),

[0036] In the formula: Represents the loss function; Indicates the true value (the degree of material degradation); Indicates the model's predicted value; Indicates the number of samples;

[0037] Convolutional neural networks are used to extract features and identify defects from images or spatial distribution data of cracks on material surfaces.

[0038] Furthermore, the formula for the repair efficiency described in step 3 is as follows:

[0039] (8),

[0040] In the formula: express Real-time repair efficiency, in percentages (%) This represents the initial repair efficiency, which is usually set to 1; This represents the repair rate coefficient, measured in units of 1 / year, and is related to environmental factors and material properties. Indicates the repair time, in years;

[0041] The optimization objective function is:

[0042] (9),

[0043] In the formula: The objective function for the repair decision represents the overall objective to be optimized during the repair process, which is usually achieved by taking multiple factors into account through weighted summation. This represents the total number of damage types; Indicates the first Weighting coefficients for different damage types; Indicates the first Changes in compressive strength for different damage types; Indicates the first Strain variation weighting coefficients for different damage types; Indicates the first Strain changes for each damage type, expressed in % . Indicates the first Weighting coefficients for repair costs of different damage types; Indicates the first Repair costs for different types of damage.

[0044] Furthermore, the optimization algorithm described in step 5 is a particle swarm optimization algorithm or a genetic algorithm;

[0045] The formula for the particle swarm optimization algorithm is as follows:

[0046] (10)

[0047] (11),

[0048] In the formula: Indicates the first The particle in the first The speed of each iteration; Indicates the first The particle in the first The position of the next iteration; Indicates inertial weight, which controls the inertia of particles; , It represents the acceleration constant, which controls the tendency of particles towards local and global optima; , Represents a random number, used to introduce randomness; Indicates the first The historical optimal solution for each particle; This represents the global optimal solution for all particles; Indicates the first The velocity of a particle in the kth iteration; Indicates the first The position of the particle in the kth iteration;

[0049] The genetic algorithm evaluates the merits of each repair strategy using a fitness function, thereby guiding the algorithm to find the optimal solution. The genetic algorithm consists of the following steps:

[0050] Step 5.3.1, Selection Operation: Select individuals with higher fitness as parents, with selection probability. It is directly proportional to the individual's fitness value, as shown in the following formula:

[0051] (12)

[0052] In the formula: Indicates the first The probability of an individual's choice; Indicates the first The fitness value of an individual is usually the value of the objective function; This represents the population size, and the total number of individuals in the population.

[0053] Step 5.3.2, Crossover Operation: Parent individuals generate new offspring individuals through a crossover operation. The crossover individuals... The formula is as follows:

[0054] (13)

[0055] In the formula: This represents the offspring individuals generated after the crossover operation; and Indicates the selected parent individual;

[0056] Step 5.3.3, Mutation Operation: Mutation is performed on the offspring individuals after crossover to increase the diversity of solutions. The formula is as follows:

[0057] (14)

[0058] In the formula: Indicates the mutated individual; This represents the offspring individuals generated after the crossover operation;

[0059] Step 5.3.4, Fitness Function: To evaluate the merits of each individual (repair strategy), a fitness function is defined. The goal is to minimize repair time and cost while maximizing repair efficiency.

[0060] (15)

[0061] In the formula: Represents an individual fitness value; , , The weight coefficients of each optimization objective in the repair strategy are used to balance the influence of different objectives; Repair time: the time required to perform the repair; Repair cost: the cost required for the repair; Repair efficiency: the efficiency of material strength recovery after repair.

[0062] Step 5.3.5, Update Operation: After each generation iteration, the population is updated through a replacement operation, selecting individuals with higher fitness to ensure the population evolves towards the optimal solution.

[0063] (16)

[0064] In the formula: Indicates the updated population; This refers to a subset of individuals from an old population that have undergone selection, crossover, and mutation. It represents the new individual after the mutation.

[0065] Furthermore, the formula for calculating the health index mentioned in step 5 is as follows:

[0066] (17)

[0067] In the formula: for The health index of a material at any given time represents the ratio of its current strength to its initial strength. for The compressive strength of the material at any given time is obtained from the degradation prediction model; Indicates the initial compressive strength of the material;

[0068] The formula for calculating the remaining lifetime is as follows:

[0069] (18)

[0070] In the formula: Indicates remaining life expectancy, in years; The degradation coefficient describes the rate of material degradation under specific conditions.

[0071] A multi-environment coupled degradation prediction and adaptive repair collaborative optimization system includes:

[0072] The first sensor network is deployed on the surface of the target structure to collect environmental factor data and material performance data, and output them to the degradation prediction modeling module.

[0073] The degradation prediction modeling module is connected to the first sensor network and is used to construct a degradation prediction model containing degradation coefficients and degradation indices to be optimized based on the environmental factor data and material performance data. The degradation coefficients and degradation indices are globally optimized using the quantum annealing algorithm, and an environmental performance nonlinear mapping is established using an artificial neural network to obtain a high-precision degradation prediction model.

[0074] A cloud-based big data platform is used to store the high-precision degradation prediction model, the repair efficiency model, and the optimal repair strategy.

[0075] The second sensor network, integrated with the adaptive repair material, is embedded inside the target structure to collect data on crack width, strain value, corrosion rate, and post-repair strength recovery in real time, and uploads it to the cloud-based big data platform.

[0076] An adaptive repair material is used to release a repair agent by rupturing microcapsules when a signal from a second sensor network indicates that the crack width or strength loss exceeds a preset threshold, thereby achieving self-repair.

[0077] The repair strategy optimization module is deployed on the cloud big data platform. It is used to calculate the repair efficiency based on the repair strength recovery data and generate the optimal repair strategy by combining repair cost, strength change, strain change and number of damage points using a multi-objective optimization algorithm.

[0078] The simulation and life assessment module, deployed on the cloud big data platform, is used to call the high-precision degradation prediction model and repair efficiency model, input real-time environmental factor data, material performance data and cumulative degradation history data, predict the material life and performance degradation trend, and evaluate the effect of each repair scheme on strength recovery and life extension.

[0079] The closed-loop update module is used to incrementally update the degradation coefficient and degradation index in the high-precision degradation prediction model by taking the strength recovery data, strain value and crack width after each repair as new samples, and driving the repair strategy optimization module to iteratively optimize until the health index is greater than or equal to the preset threshold and the remaining life is greater than or equal to the preset life value.

[0080] Furthermore, it also includes an API interface module for outputting the optimal repair strategy to an external building management system, construction management system, or health monitoring system.

[0081] Furthermore, the adaptive repair material includes a self-healing polymer matrix, a repair agent, and a reinforcing material. The self-healing polymer matrix is ​​epoxy resin or polyurethane resin, the repair agent is embedded in the self-healing polymer matrix in the form of microcapsules or fiber mesh, and the reinforcing material is nanomaterials such as carbon nanotubes or graphene.

[0082] A computer program product comprising a storage medium and computer-readable instructions stored on the medium, which, when executed by a computer, implement the method described.

[0083] Advantages of the present invention

[0084] 1. The multi-environment coupled degradation prediction and adaptive repair collaborative optimization method and system of the present invention adopts a degradation prediction model coupled with multiple environmental loads and materials, comprehensively considers the influence of multiple factors such as temperature, humidity, salinity, and pressure on the performance of target structural materials, and combines quantum computing and artificial intelligence optimization to significantly improve the accuracy of degradation prediction.

[0085] 2. This invention combines adaptive repair materials with an intelligent monitoring system, enabling real-time monitoring of the material's health status and automatic repair upon detection of cracks or damage. This avoids continued material degradation caused by delayed repair in traditional methods. Combined with optimized repair strategies, the repair plan can be dynamically adjusted according to different environments and material conditions, improving the timeliness and effectiveness of repairs.

[0086] 3. The multi-environment collaborative simulation platform designed in this invention can simulate the material degradation process in extreme environments such as deep sea, polar regions, the moon, and Mars bases. It can perform material health management and repair optimization under special environments such as high pressure, high temperature, radiation, and extremely low temperature. It has strong scalability and is applicable to a variety of materials and working conditions.

[0087] 4. This invention combines material health monitoring, repair decision-making, and feedback mechanisms, integrating a monitoring-prediction-repair-feedback closed loop. It enables real-time monitoring, evaluation, and optimization during material use, and dynamically adjusts repair strategies through real-time feedback to ensure continuous optimization of repair effects, greatly improving the long-term stability and safety of the material. It also reduces human intervention, avoids excessive or delayed repairs, and maximizes cost-effectiveness.

[0088] 5. This invention can seamlessly integrate with building management systems (BMS), construction management systems, and health monitoring systems to achieve cross-platform collaboration, ensuring intelligent collaborative work at every stage from material health monitoring to repair execution, thereby improving overall operation and maintenance efficiency. Attached Figure Description

[0089] Figure 1 This is a flowchart of the multi-environment coupled degradation prediction and adaptive repair collaborative optimization method of the present invention.

[0090] Figure 2 for Figure 1 The flowchart for the big data closed-loop optimization, repair, feedback, and update process in step 6. Detailed Implementation

[0091] The present invention will be further explained and described below with reference to the accompanying drawings and specific embodiments. It should be noted that the specific embodiments are not intended to limit the scope of the present invention.

[0092] like Figure 1 and Figure 2 As shown in the figure, the multi-environment coupled degradation prediction and adaptive repair collaborative optimization method provided in this specific embodiment includes the following steps:

[0093] Step 1: By deploying a first sensor network on the surface of the target structure to collect environmental factor data and material performance data, a degradation prediction model coupled with multiple environmental loads and including the degradation coefficient and degradation index to be optimized is constructed. The environmental factor data includes temperature, humidity, salinity, and pressure, and the material performance data includes initial compressive strength, elastic modulus, coefficient of thermal expansion, and porosity. The specific steps are as follows:

[0094] Step 1.1, Environmental Data Collection and Processing: Environmental factor data is collected through environmental sensors, weather stations, satellite remote sensing, etc. This data includes four environmental factors: temperature, humidity, salinity, and pressure. These four factors collectively influence the material degradation process. The variation range of each environmental factor and its impact on material properties will serve as input to the degradation prediction model. The variation data of each environmental factor in spatiotemporal coordinates are directly used as input to the degradation prediction model coupled with multiple environmental loads.

[0095] temperature( ( ): The unit is ℃, which represents the change in ambient temperature. Temperature difference has a significant impact on the thermal expansion and stress change of materials.

[0096] humidity( ): The unit is %, which describes the effect of environmental humidity on materials, especially the effect on the corrosion rate of materials in corrosive environments.

[0097] salinity( : The unit is ppt (parts per thousand), used to describe the concentration of salt water in the environment, which affects the corrosion rate and crack propagation of materials.

[0098] pressure( Pa (Pascal): This unit describes environmental pressure, especially in deep-sea or underground environments, and affects the compressibility and strength of materials.

[0099] Step 1.2, Acquisition of Material Performance Data: Material performance data quantifies the degradation behavior and performance changes of the target structural material under different environments. Material performance data includes compressive strength, elastic modulus, coefficient of thermal expansion, and porosity. Compressive strength and elastic modulus are obtained through standard testing methods (such as compression testing). The coefficient of thermal expansion and porosity are determined through experimental testing and chemical analysis.

[0100] Compressive strength ( ): The unit is MPa, which represents the compressive strength of a material and is the primary parameter for material degradation.

[0101] elastic modulus ( ): The unit is MPa, which describes the elastic deformation capacity of a material under load.

[0102] coefficient of thermal expansion ( ): The unit is 1 / ℃, which describes the volume change of a material due to temperature changes.

[0103] Porosity ( (%): The unit is %, which describes the proportion of pores inside a material and affects the material's permeability and corrosion resistance.

[0104] Step 1.3: Construct a degradation prediction model for materials coupled under multiple environmental loads: A degradation prediction model for materials coupled under multiple environmental loads is established by combining the environmental factor data from Step 1.1 and the material performance data from Step 1.2. This degradation prediction model can predict the material performance degradation process under different environmental conditions and provide a basis for subsequent repair strategies.

[0105] The formula for constructing the environmental load-material coupled degradation model is as follows:

[0106] (1),

[0107] In the formula: express The compressive strength of the material at all times; Indicates the initial compressive strength; This function represents the influence of environmental factors on material properties, specifically the effects of temperature, humidity, salinity, and pressure. It is fitted to environmental data to describe the degradation effects of these factors on the material. The degradation coefficient, expressed in units of 1 / year, describes the rate of material degradation under harsh environments and depends on environmental factors and the type of material; the degradation rate under different environments is obtained through experiments. Indicates time, in years, indicating the service life of the material; The degradation index represents the degree of acceleration of material degradation; this degradation index is obtained by fitting experimental data under different material and environmental conditions. The values ​​are different.

[0108] Step 1.4, Formula for the Impact of Environmental Factor Data on Material Degradation: The specific impact of environmental factor data needs to be fitted using experimental or empirical data. The following is a function describing the impact of environmental factor data:

[0109] (2),

[0110] In the formula: , , , , These are empirical constants and need to be fitted using experimental data; Indicates temperature; Indicates humidity; Indicates salinity; Indicates pressure; e bT The exponential function represents the accelerating effect of temperature on degradation; ln(S) represents the logarithmic function of the effect of salinity.

[0111] The function representing the influence of environmental factors indicates the combined effect of environmental factors on the performance of the target structural material. Temperature, humidity, salinity, and pressure under different environmental conditions will have a nonlinear effect on the compressive strength, crack propagation, corrosion rate, and other properties of the material.

[0112] The effect of temperature (T): Changes in temperature cause thermal expansion of materials, leading to changes in stress, which in turn affects the strength of the materials. Increased temperature accelerates the degradation process of materials.

[0113] The effect of humidity (H): Increased humidity usually increases the corrosion rate of materials, especially for metallic materials.

[0114] Effects of salinity (S): Increased salinity accelerates the electrochemical corrosion of materials, especially metallic materials.

[0115] Effects of pressure (P): Under high pressure, the compressibility and strength of materials are significantly affected, especially in deep-sea and underground environments.

[0116] Step 1.5, Experimental verification and model optimization of the degradation process

[0117] Validation of experimental data: In order to ensure the accuracy of the degradation prediction model of materials coupled with multiple environmental loads, it is necessary to test the degradation process of materials under different environmental conditions in the experiment, record the degradation rate and strength changes under different environments, and optimize the parameters in the model based on the experimental data.

[0118] Model optimization: By using methods such as least squares or maximum likelihood estimation, the experimental data is fitted to obtain the optimal parameters, thereby further improving the prediction accuracy.

[0119] Step 2: Using the environmental factor data (temperature, humidity, salinity, pressure) and material performance data (compressive strength, elastic modulus, porosity) collected in Step 1 as the original samples, and combining them with the spatiotemporal degradation history data corresponding to the original samples, a standardized training dataset is constructed after missing value processing and standardization. The degradation coefficient and degradation index to be optimized in the degradation prediction model of Step 1 are globally optimized using the quantum annealing algorithm. Then, the optimized degradation coefficient and degradation index are fixed, and an artificial neural network is used to establish a nonlinear mapping of environmental performance to obtain a high-precision degradation prediction model. The degradation history data includes the degradation rate, crack propagation, and strength loss of materials under different environments.

[0120] The specific steps are as follows:

[0121] Step 2.1, Data Processing Method:

[0122] (1) Standardization and Normalization: To avoid bias in model training caused by data from different dimensions, it is necessary to standardize (or normalize) the original samples and degraded historical data. The following standardization formula is used for each environmental factor data and material performance data:

[0123] (3),

[0124] In the formula: Represents the original data value; This represents the data mean; Indicates the standard deviation of the data;

[0125] (2) Data cleaning: Remove missing and outlier values. Fill in missing data using interpolation or mean substitution to ensure data integrity.

[0126] (3) Feature selection: Use feature selection algorithms (such as principal component analysis (PCA) or information gain) to select the features that have the greatest impact on the prediction results and reduce the data dimensionality.

[0127] Step 2.2, Quantum Annealing is a quantum computing technique used to solve optimization problems. It finds the optimal solution in a multidimensional parameter space, greatly accelerating the parameter optimization process. It is particularly suitable for the parameter optimization problem in this specific embodiment. In the degradation prediction model of the target structural material, the quantum annealing algorithm is used to optimize the degradation coefficient (…). ) and degradation index ( This is done to maximize the model's prediction accuracy. The formula for the quantum degradation algorithm is as follows:

[0128] (4),

[0129] In the formula: Corresponding energy state The probability of; It represents the energy state, specifically the optimization target of the degradation coefficient and degradation index; Represents the Boltzmann constant; This represents the temperature change during quantum annealing; represents the partition function, used to normalize the probability distribution; e represents the base of the natural logarithm, which is 2.71828, used for exponential calculation;

[0130] Traditional classical optimization algorithms, such as gradient descent or genetic algorithms, may face computational efficiency bottlenecks on large-scale datasets. Compared to classical optimization algorithms, this specific embodiment uses quantum annealing for parallel computation and quantum superposition states, significantly improving computational speed.

[0131] For example, traditional genetic algorithms may require traversing a large solution space, while quantum computing can utilize quantum superposition to consider multiple solutions simultaneously, thereby accelerating the optimization process.

[0132] Step 2.3: After obtaining the optimized degradation prediction model parameters, an artificial neural network is selected to optimize the degradation prediction model. Deep learning can handle nonlinear complex relationships, thus providing more accurate prediction results. The artificial neural network is a deep neural network or a convolutional neural network. The deep neural network is used to perform regression prediction on the compressive strength and degradation rate of materials. When training the deep neural network, the goal is to minimize the loss function so that the model's prediction results are as close as possible to the actual data. The commonly used loss function is the mean squared error (MSE), and the formula is as follows:

[0133] (5),

[0134] In the formula: Represents the loss function; Indicates the true value (the degree of material degradation); Indicates the model's predicted value; Indicates the number of samples;

[0135] Convolutional neural networks are used to extract features and identify defects from images or spatial distribution data of cracks on material surfaces.

[0136] The model is trained using the Adam or RMSprop optimizer, and the learning rate and other hyperparameters are adjusted. The prediction accuracy of the degradation prediction model is further improved through cross-validation and hyperparameter tuning.

[0137] Step 2.4: To verify the effectiveness of the trained high-precision degradation prediction model, cross-validation and a test set are used to evaluate its predictive ability. By comparing the model's prediction results with real experimental data, metrics such as mean squared error (MSE) and mean absolute error (MAE) are calculated to evaluate the deviation between the model's predicted values ​​and the actual values, thereby measuring the model's accuracy.

[0138] The formula for the mean squared error (MSE) is as follows:

[0139] (6),

[0140] The formula for the mean absolute error (MAE) is as follows:

[0141] (7),

[0142] Coefficient of determination ( : measures the goodness of fit of the model; a value close to 1 indicates strong predictive ability.

[0143] Equations (6) and (7), where N represents the sample size; y i Represents the true value, which is the first... The actual degradation status of each sample; This represents the predicted value, which is the first value predicted by the model. The degradation status of each sample;

[0144] By testing in different environments and with different materials, we ensure that the high-precision degradation prediction model has high prediction accuracy under various extreme conditions.

[0145] Step 3: Embed adaptive repair material and deploy a second sensor network in the target structure. The adaptive repair material includes a self-healing polymer matrix, a repair agent, and a reinforcing material. The self-healing polymer matrix is ​​epoxy resin or polyurethane resin, which has high adhesion and good corrosion resistance. The repair agent is embedded in the self-healing polymer matrix in the form of microcapsules or fiber mesh. When the material is damaged, the microcapsules rupture, releasing the repair agent and automatically repairing the cracks. The reinforcing material is a nanomaterial such as carbon nanotubes or graphene, which can enhance the mechanical strength and wear resistance of the material. The adaptive repair material can automatically repair microcracks and damage caused by environmental changes, stress, and other factors, extending the service life of the material. The second sensor network is integrated with the adaptive repair material and embedded in real time to collect crack width, strain value and corrosion rate as health data. When the second sensor network detects cracks or strength loss exceeding a threshold, the microcapsules rupture and release the repair agent. At the same time, it continuously collects post-repair strength recovery data and transmits it back to the cloud big data platform. The repair efficiency is calculated based on the post-repair strength recovery data. Combined with repair cost, strength change, strain change and number of damage points, an initial repair strategy is generated using an optimization objective function.

[0146] Step 3.1, Damage Triggering and Self-Healing: When microcracks or damage occur in the target structural material, the cracks trigger the rupture of microcapsules inside the self-healing polymer, releasing a repair agent (such as epoxy resin repair fluid). The repair agent fills the crack and, through a chemical reaction, bonds with the material on both sides of the crack, restoring the material's original strength.

[0147] The formula for the repair efficiency is as follows:

[0148] (8),

[0149] In the formula: express Real-time repair efficiency, in percentages (%) This represents the initial repair efficiency, which is usually set to 1; This represents the repair rate coefficient, measured in units of 1 / year, and is related to environmental factors and material properties. Indicates the repair time, in years;

[0150] Repair duration: The effectiveness and duration of the repair agent will gradually decrease over time, so the durability of the repair material is an important design parameter.

[0151] Step 3.2, Intelligent Monitoring System and Cloud Big Data Platform: The intelligent monitoring system includes a first sensor network deployed on the surface and a second sensor network embedded inside the structure. Both the first and second sensor networks integrate strain sensors, corrosion sensors, and temperature and humidity sensors to collect environmental factor data and material performance data in real time. The system monitors the development of cracks, deformation, and environmental changes on the material surface in real time and provides a basis for subsequent repair decisions through the cloud big data platform.

[0152] A strain gauge measures the strain on a material surface and can detect the formation and propagation of cracks.

[0153] Corrosion sensors monitor the corrosion status of materials in real time, especially in saltwater environments or extreme climatic conditions, and the impact of corrosion on repair materials.

[0154] Temperature and humidity sensors are used to detect changes in the temperature and humidity of the environment in which a material is located, thereby affecting the material's self-healing efficiency.

[0155] Cloud-based big data platform: Sensor data is transmitted to the cloud-based big data platform in real time via wireless sensor networks (WSN) for easy monitoring and storage.

[0156] The data includes multiple dimensions such as strain, corrosion degree, crack width, and environmental changes.

[0157] Data storage and analysis: Through big data storage and machine learning algorithms on a cloud-based big data platform, monitoring data is stored, processed, and analyzed to generate health reports for the materials.

[0158] The data flow and transmission process are as follows:

[0159] (1) Data collection: The sensor node, consisting of strain sensor, corrosion sensor and temperature and humidity sensor, collects crack width, strain, corrosion rate and environmental parameters in real time and caches them in the local memory of the node.

[0160] (2) Data upload and storage: Sensor nodes upload data to the cloud big data platform for storage and backup through wireless communication technologies (such as LoRaWAN and NB-IoT).

[0161] (3) Data analysis and decision-making: The cloud-based big data platform analyzes health data through machine learning models, and provides repair suggestions and health assessments.

[0162] Step 3.3, Decision-making and Optimization of Repair Strategy: Based on the data collected by the intelligent monitoring system, the health status of the materials can be automatically analyzed and repair strategies can be generated. By introducing artificial intelligence (AI) and optimization algorithms, the repair strategy is dynamically adjusted to achieve the best results under different environmental conditions.

[0163] (1) Optimization formula for the repair strategy:

[0164] In multi-environment and multi-damage scenarios, the repair strategy is determined by the following optimization objective function:

[0165] The optimization objective function is:

[0166] (9),

[0167] In the formula: The objective function for the repair decision represents the overall objective to be optimized during the repair process, which is usually achieved by taking multiple factors into account through weighted summation. This represents the total number of damage types; Indicates the first Weighting coefficients for different damage types; Indicates the first Changes in compressive strength for different damage types; Indicates the first Strain variation weighting coefficients for different damage types; Indicates the first Strain changes for each damage type, expressed in % . Indicates the first Weighting coefficients for repair costs of different damage types; Indicates the first Repair costs for different types of damage.

[0168] (2) Intelligent Repair Decision System:

[0169] After collecting sensor data, the effectiveness of the current repair strategy is calculated using optimization algorithms, and repair time, methods, and materials to be used are recommended. The results of each repair are fed back to a cloud-based big data platform, and the strategy is continuously adjusted and optimized based on the feedback.

[0170] Step 3.4, Collaborative Operation of Adaptive Repair Material and Intelligent Monitoring System: The adaptive repair material and intelligent monitoring system work together to form a closed-loop control system. When the intelligent monitoring system detects cracks or a decline in material performance, the repair material automatically repairs the cracks, and the repair effect is monitored through a first sensor network and a second sensor network. The repair strategy is adjusted based on real-time data to ensure the timeliness and effectiveness of the repair. The repair efficiency of the repair material varies with time and environmental conditions; therefore, the repair strategy needs to be adjusted in real time. The release amount of the repair material and the repair time are adjusted based on real-time feedback of the material's health data and environmental factor data from the intelligent monitoring system.

[0171] The data from each repair will serve as input for a new round of optimization, ensuring the long-term performance stability of the material.

[0172] Step 4: Write the high-precision degradation prediction model from Step 2 and the repair efficiency from Step 3 into the simulation platform. Input real-time environmental factor data, material performance data, and historical degradation data accumulated from Step 3 to predict the material life and performance degradation trend. Based on the repair strategy generated in Step 3, evaluate the effect of each repair scheme on material strength recovery and life extension to obtain life prediction results and repair effect evaluation results. Use optimization algorithms to perform multi-objective optimization on repair time and repair material type to form the optimal repair strategy.

[0173] Step 4.1, the simulation platform includes the following core modules:

[0174] (1) Environmental Factor Simulation Module: Simulates the changes in different environmental conditions (such as temperature, humidity, salinity, and pressure) and their effects on material degradation.

[0175] (2) Material degradation simulation module: Based on a high-precision degradation prediction model, it simulates the performance degradation, crack propagation, corrosion rate, etc. of the target structural material under different environmental conditions.

[0176] (3) Repair and optimization decision module: Based on the simulation results, it automatically provides repair strategies and lifetime predictions, and adjusts the repair plan through a feedback mechanism.

[0177] (4) Data storage and processing module: used to store experimental data, simulation results and monitoring data, and to perform big data analysis and mining.

[0178] Step 4.1.2, enter the following data:

[0179] Environmental factor data: such as temperature ( ),humidity( ),salinity( ),pressure( )wait.

[0180] Material performance data: such as compressive strength ( ), elasticity model ( )wait.

[0181] Degradation history data: such as crack propagation, loss of compressive strength, etc.

[0182] Simulation process:

[0183] 1. Input environmental factors and material performance data.

[0184] 2. The degradation process of materials is predicted by using a high-precision degradation prediction model.

[0185] 3. Make repair and optimization decisions and simulate the impact of different repair strategies on material life.

[0186] 4. Output a health assessment report, including the remaining lifespan of the materials and repair recommendations.

[0187] Step 4.2: During the simulation process, the established high-precision degradation prediction model is written into the simulation platform, and the complex coupling relationship between environmental factors and material degradation is realized through simulation algorithms. This enables the high-precision degradation prediction model to not only simulate degradation processes under common environments, but also to handle changes in material properties under extreme environmental conditions.

[0188] The simulation formula for the high-precision degradation prediction model is as follows:

[0189] As described in step one, the compressive strength of the material Over time The change is calculated using the following formula:

[0190] (1),

[0191] In the formula: express The compressive strength of the material at all times; Indicates the initial compressive strength; This function represents the influence of environmental factors on material properties, including the effects of factors such as temperature, humidity, salinity, and pressure. This function is obtained through regression analysis or fitting experimental data. The degradation coefficient (unit: 1 / year) represents the rate of material degradation under harsh environments; Indicates time (unit: year); The degradation index indicates the rate of material degradation.

[0192] The simulation steps of the high-precision degradation prediction model are as follows:

[0193] (1) Environmental factor input: Input different environmental conditions (such as high temperature, high humidity, high salinity, etc.) into the simulation platform.

[0194] (2) Material property calculation: Based on the initial properties of the material and environmental conditions, the mechanical properties of the material under different environments are calculated using a degradation model.

[0195] (3) Degradation prediction: Use a high-precision degradation prediction model to predict the life and performance degradation trend of materials and output the prediction results.

[0196] (4) Evaluation of repair effect: Through repair strategy simulation, evaluate the effect of different repair schemes on material strength recovery and life extension.

[0197] Step 4.3, Repair and Optimization Decision Module: After the simulation of the target structure material degradation is completed, the repair and optimization decision module will propose a repair plan based on the simulation results. This module optimizes the repair strategy using artificial intelligence algorithms, making the repair process more accurate and efficient.

[0198] The optimization objective of the repair strategy optimization module is to minimize repair costs, maximize material lifespan, and improve repair efficiency. The objective function is:

[0199] (9),

[0200] In the formula: The objective function for the repair decision represents the overall objective to be optimized during the repair process, which is usually achieved by taking multiple factors into account through weighted summation. This represents the total number of damage types; Indicates the first Weighting coefficients for different damage types; Indicates the first Changes in compressive strength for different damage types; Indicates the first Strain variation weighting coefficients for different damage types; Indicates the first Strain changes for each damage type, expressed in % . Indicates the first Weighting coefficients for repair costs of different damage types; Indicates the first Repair costs for different types of damage.

[0201] Optimization algorithms such as Genetic Algorithm (GA) or Particle Swarm Optimization (PSO) are used to minimize the above objective function to obtain the optimal repair scheme. The repair decision is determined by the algorithm:

[0202] Repair time: Select the optimal time for repair to minimize the impact of the repair cycle on structural performance.

[0203] Repair material selection: Choose the most suitable repair material based on the extent of damage and environmental conditions.

[0204] Step 4.4, Data Storage and Analysis Module: All simulation and monitoring data need to be stored and processed for subsequent health assessments and decision optimization. Data is stored and analyzed in real-time through a cloud-based big data platform. Data storage utilizes distributed databases (such as Hadoop and Spark), supporting efficient storage and retrieval of massive amounts of data. Data mining techniques are used to analyze the degradation patterns of materials under different environments, providing a basis for remediation decisions. Machine learning models and predictive algorithms are used to evaluate the remediation effect and make optimizations based on the remediation data.

[0205] Step 5: Perform repair on the target structure according to the optimal repair strategy obtained in Step 4, and feed the repair effect back to the high-precision degradation prediction model in Step 2 for dynamic updating of model parameters. Using repair cost, strength change, strain change, and number of damage points as optimization objectives, iteratively optimize the repair time, repair materials, and repair method using the same multi-objective optimization algorithm as in Step 4 to obtain a dynamically updated repair strategy. Calculate the health index based on the material compressive strength at time t and the initial compressive strength output by the high-precision degradation prediction model in Step 2, and calculate the remaining lifetime based on this health index and the degradation coefficient obtained in Step 2. The specific steps are as follows:

[0206] Step 5.1, the optimization of the repair strategy not only considers the current health status of the target structural material, but also needs to predict future degradation trends and formulate precise repair plans based on different damage and environmental conditions.

[0207] The goal of optimizing the repair strategy:

[0208] (1) Maximize material lifespan: By selecting appropriate repair schemes, delay the degradation process of materials to the greatest extent.

[0209] (2) Minimize repair costs: Select the most cost-effective repair method and avoid over-repair.

[0210] (3) Improve repair efficiency: Ensure the timeliness and accuracy of the repair process and reduce the impact on the overall function of the structure.

[0211] Step 5.2 establishes a repair strategy optimization model. This model aims to maximize lifespan and minimize cost, and comprehensively considers material degradation rate, environmental conditions, repair methods, and repair time.

[0212] Optimize the objective function:

[0213] (9),

[0214] In the formula: The objective function for the repair decision represents the overall objective to be optimized during the repair process, which is usually achieved by taking multiple factors into account through weighted summation. This represents the total number of damage types; Indicates the first Weighting coefficients for different damage types; Indicates the first Changes in compressive strength for different damage types; Indicates the first Strain variation weighting coefficients for different damage types; Indicates the first Strain changes for each damage type, expressed in % . Indicates the first Weighting coefficients for repair costs of different damage types; Indicates the first Repair costs for different types of damage.

[0215] Repair decision: By optimizing the objective function, the optimal repair strategy for each repair point is calculated, including repair time, selection of repair materials, and repair method.

[0216] Step 5.3: Solve the optimization objective function using intelligent optimization algorithms such as Particle Swarm Optimization (PSO) or Genetic Algorithm (GA). These algorithms can effectively search for the global optimum in a multidimensional parameter space, avoiding the problem of local optima.

[0217] The formula for the particle swarm optimization algorithm is as follows:

[0218] (10)

[0219] (11),

[0220] In the formula: Indicates the first The particle in the first The speed of each iteration; Indicates the first The particle in the first The position of the next iteration; Indicates inertial weight, which controls the inertia of particles; , It represents the acceleration constant, which controls the tendency of particles towards local and global optima; , Represents a random number, used to introduce randomness; Indicates the first The historical optimal solution for each particle; This represents the global optimal solution for all particles; Indicates the first The velocity of a particle in the kth iteration; Indicates the first The position of the particle in the k-th iteration.

[0221] The genetic algorithm evaluates the merits of each repair strategy using a fitness function, thereby guiding the algorithm to find the optimal solution. The genetic algorithm consists of the following steps:

[0222] Step 5.3.1, Selection Operation: Select individuals with higher fitness as parents, with selection probability. It is directly proportional to the individual's fitness value, as shown in the following formula:

[0223] (12)

[0224] In the formula: Indicates the first The probability of an individual's choice; Indicates the first The fitness value of an individual is usually the value of the objective function; This represents the population size, and the total number of individuals in the population.

[0225] Step 5.3.2, Crossover Operation: Parent individuals generate new offspring individuals through a crossover operation. The crossover individuals... The formula is as follows:

[0226] (13)

[0227] In the formula: This represents the offspring individuals generated after the crossover operation; and Indicates the selected parent individual;

[0228] Step 5.3.3, Mutation Operation: Mutation is performed on the offspring individuals after crossover to increase the diversity of solutions. The formula is as follows:

[0229] (14)

[0230] In the formula: Indicates the mutated individual; This represents the offspring individuals generated after the crossover operation;

[0231] Step 5.3.4, Fitness Function: To evaluate the merits of each individual (repair strategy), a fitness function is defined. The goal is to minimize repair time and cost while maximizing repair efficiency.

[0232] (15)

[0233] In the formula: Represents an individual fitness value; , , The weight coefficients of each optimization objective in the repair strategy are used to balance the influence of different objectives; Repair time: the time required to perform the repair; Repair cost: the cost required for the repair; Repair efficiency: the efficiency of material strength recovery after repair.

[0234] Step 5.3.5, Update Operation: After each generation iteration, the population is updated through a replacement operation, selecting individuals with higher fitness to ensure the population evolves towards the optimal solution.

[0235] (16)

[0236] In the formula: Indicates the updated population; This refers to a subset of individuals from an old population that have undergone selection, crossover, and mutation. It represents the new individual after the mutation.

[0237] Particle swarm optimization (PSO) or genetic algorithm (GA) can effectively find the optimal solution for the repair strategy, including the choice of repair time, repair method, and repair material.

[0238] Step 5.4: After optimizing the repair strategy, continuous monitoring and health assessment of the material throughout its entire lifecycle are also required. Lifecycle management not only helps monitor the material's health status in real time but also predicts future degradation trends, allowing for timely adjustments to the repair strategy.

[0239] 1. Key elements of lifecycle management:

[0240] Health monitoring: The intelligent monitoring system collects data on material degradation, crack propagation, corrosion rate, etc. in real time, providing data support for subsequent repair decisions.

[0241] Degradation prediction: Using a degradation prediction model optimized with quantum computing and artificial intelligence to predict the future degradation trend of materials and prepare for repair in advance.

[0242] Repair decision support: During the repair process, the repair strategy is dynamically adjusted based on real-time monitoring data and degradation prediction results to ensure that the timing and method of each repair are optimal.

[0243] 2. Health assessment formula:

[0244] Health assessment is achieved by comprehensively considering the material's remaining lifespan and health index. Health Index Calculated using the following formula:

[0245] (17)

[0246] In the formula: for The health index of a material at any given time represents the ratio of its current strength to its initial strength. for The compressive strength of the material at any given time is obtained from the degradation prediction model; This indicates the initial compressive strength of the material.

[0247] 3. Life expectancy prediction: combined with health index The degradation rate of materials can be used to predict the remaining life of the target structural material. Remaining life The calculation formula is:

[0248] (18)

[0249] In the formula: Indicates remaining life expectancy (unit: years); The degradation coefficient describes the rate of material degradation under specific conditions.

[0250] Step 5.5: Dynamically adjust the repair strategy based on real-time monitoring data and repair results through the feedback mechanism.

[0251] Feedback mechanism process:

[0252] 1. Data Acquisition: The material condition, including crack propagation, corrosion degree, stress change, etc., is monitored in real time through the first sensor network and the second sensor network.

[0253] 2. Data transmission and storage: Transmit data to a cloud-based big data platform for storage and analysis.

[0254] 3. Repair effectiveness assessment: Based on the health assessment of the target structural materials, evaluate the effectiveness of the current repair strategy and determine whether further repair is needed.

[0255] 4. Adjustment of remediation strategy: Dynamically adjust the remediation strategy based on the assessment results and the predicted data in lifecycle management.

[0256] This feedback mechanism ensures that repair strategies can be optimized in a timely manner during material degradation, thus ensuring the stability and efficiency of the material throughout its entire life cycle.

[0257] Step 6, closed-loop optimization and repair feedback update based on big data, includes the following sub-steps:

[0258] Step 6.1: Upload the high-precision degradation prediction model obtained in Step 2, the repair efficiency obtained in Step 3, and the optimal repair strategy obtained in Step 4 to the cloud big data platform.

[0259] Step 6.2: Continuously acquire environmental factor data, material performance data, crack width, strain value, corrosion rate, and post-repair strength recovery data through the first sensor network in Step 1 and the second sensor network in Step 3. After data cleaning, missing value filling, and normalization, new samples are formed.

[0260] Environmental factor data: Environmental factors such as temperature, humidity, salinity, and pressure are collected in real time through the first and second sensor networks to provide input for the degradation prediction model.

[0261] Material condition data: Data such as crack propagation, compressive strength, and deformation of materials are collected through devices such as strain sensors and corrosion sensors.

[0262] Post-repair strength recovery data: Data after each repair, including information such as the type of repair material, repair time, and repair effect.

[0263] Real-time monitoring data: Monitor material temperature changes, crack conditions, corrosion levels, etc., and provide real-time feedback on the material's health status.

[0264] Data cleaning and standardization: The collected data is denoised, missing values ​​are removed, and the data is standardized to ensure consistency and comparability.

[0265] Feature selection: Use PCA (principal component analysis) or information gain methods to select important features related to material degradation and repair effectiveness, and reduce redundant information.

[0266] Data integration: Integrating data from different sources (such as environmental data, material data, and remediation data) to build a complete big data platform, providing support for subsequent analysis and model optimization.

[0267] The data storage and management

[0268] Data is stored through a cloud-based big data platform, supporting distributed storage and efficient querying of big data, ensuring data security and high availability.

[0269] Use NoSQL databases (such as MongoDB and Cassandra) or distributed file systems (such as Hadoop HDFS) to store and manage massive amounts of data.

[0270] Step 6.3: Use the new samples from Step 6.2 to incrementally update the degradation coefficients and degradation indices that were fixed in Step 2, and then send the updated degradation coefficients and degradation indices back to the high-precision degradation prediction model from Step 2 to dynamically refresh the model parameters.

[0271] After data collection and storage are completed, the high-precision degradation prediction model in step 2 is used to predict the future degradation process of materials using historical data, ensuring timely repair and preventing materials from reaching an irreparable state during the degradation process.

[0272] The degradation prediction formula is as follows:

[0273] (1),

[0274] In the formula: Indicates the material is in The compressive strength at any given time; Indicates the initial compressive strength; A function representing the influence of environmental factors on material properties; The degradation coefficient represents the rate of material degradation under specific environmental conditions. The degradation index indicates the degree of acceleration of material degradation; Indicates time (unit: year).

[0275] Machine learning algorithms are used to optimize repair strategies based on historical data. Regression analysis or deep learning are used to model key factors affecting repair effectiveness (such as repair time, material selection, and environmental conditions) and calculate the optimal repair strategy.

[0276] The optimization objective of the repair strategy is calculated using the following objective function:

[0277] (9),

[0278] In the formula: Let represent the optimization objective of the repair decision, and let represent the overall optimization objective during the repair process. , This represents the influence coefficient of different damage types (such as crack propagation, strength loss, etc.) on repair decisions during the repair process; This indicates the change in the compressive strength of the material after repair, reflecting the impact of repair on the recovery of the material's strength. The value represents the strain change of the material after repair (unit: %), reflecting the impact of repair on the material's deformation recovery. The repair cost (unit: yuan) includes material costs, labor costs, etc. This indicates the number of material damage points and represents the cumulative effect under multiple damage conditions.

[0279] The objective function is solved using intelligent optimization methods such as genetic algorithm (GA) or particle swarm optimization (PSO) to determine the optimal repair strategy.

[0280] Step 6.4, Repair Feedback and Further Optimization:

[0281] 1. Feedback Mechanism Process

[0282] Post-repair data acquisition: After each repair, the first and second sensor networks collect data on the repair effect, changes in material properties, and environmental changes in real time, and feed them back to the cloud big data platform.

[0283] Data Analysis and Evaluation: The cloud-based big data platform analyzes the collected feedback data to assess the repair effectiveness and whether the expected goals have been achieved. If the repair effectiveness does not meet expectations, the cloud-based big data platform will trigger adjustments to the repair strategy.

[0284] Dynamic adjustment of repair strategies: Based on feedback data and degradation prediction results, the cloud-based big data platform will automatically optimize repair time and methods, and generate new repair solutions to address potential material degradation and environmental changes.

[0285] The core of the feedback mechanism lies in the evaluation of repair effectiveness and strategy adjustment. The compressive strength after repair ( ) and strain ( The following feedback adjustment formula will be used for evaluation:

[0286] (19)

[0287] (20)

[0288] In the formula: Indicates the compressive strength after repair; Indicates the strain after repair; This indicates the change in material strength during the repair process; This indicates the change in material strain during the repair process.

[0289] If the repair effect is not ideal, the system will adjust the optimization strategy and calculate a new repair plan in order to restore the material's performance in a timely manner.

[0290] 2. Input the refreshed high-precision degradation prediction model back into the simulation platform of step 4. Combine it with the strength recovery data after repair accumulated in step 6.2, recalculate the repair efficiency, and use the same multi-objective optimization algorithm as in step 4 to iteratively optimize the repair time and the type of repair material. Use the optimization result as the new current optimal repair strategy.

[0291] Optimized design and a repair feedback mechanism work together to provide a closed-loop control system for the material through data analysis and optimization algorithms. The optimized design first selects the optimal repair strategy based on data prediction. After repair, feedback data is obtained through a monitoring system, allowing for timely adjustments to the repair strategy and ensuring an efficient and continuously optimized repair process. Throughout the entire service life of the target structural material, the repair strategy is continuously optimized and adjusted, with each repair based on the latest degradation prediction and feedback data. This ensures the material remains in optimal health and maximizes its service life.

[0292] Step 6.5: After performing the repair according to the new current optimal repair strategy in Step 6.4, continue to collect the strength recovery data, strain value and crack width after repair through the second sensor network. Repeat steps 6.2 to 6.4 with the newly added data to obtain the updated current optimal repair strategy.

[0293] Step 6.6: Repeat steps 6.2 to 6.5 to make the high-precision degradation prediction model in step 2 converge continuously with the measured data until the health index output in step 2 is greater than or equal to the preset threshold and the remaining lifespan is greater than or equal to the preset lifespan value.

[0294] A computer program product comprising a storage medium and computer-readable instructions stored on the medium, which, when executed by a computer, implement the method described above.

[0295] This specific embodiment also provides a multi-environment coupled degradation prediction and adaptive repair collaborative optimization system, including:

[0296] The first sensor network is deployed on the surface of the target structure to collect environmental factor data and material performance data, and output them to the degradation prediction modeling module.

[0297] The degradation prediction modeling module is connected to the first sensor network and is used to construct a degradation prediction model containing degradation coefficients and degradation indices to be optimized based on the environmental factor data and material performance data. The degradation coefficients and degradation indices are globally optimized using the quantum annealing algorithm, and an environmental performance nonlinear mapping is established using an artificial neural network to obtain a high-precision degradation prediction model.

[0298] A cloud-based big data platform is used to store the high-precision degradation prediction model, the repair efficiency model, and the optimal repair strategy.

[0299] The second sensor network, integrated with the adaptive repair material, is embedded inside the target structure to collect data on crack width, strain value, corrosion rate, and post-repair strength recovery in real time, and uploads it to the cloud-based big data platform.

[0300] An adaptive repair material is provided, comprising a self-healing polymer matrix, a repair agent, and a reinforcing material. The self-healing polymer matrix is ​​epoxy resin or polyurethane resin, the repair agent is embedded in the self-healing polymer matrix in the form of microcapsules or fiber meshes, and the reinforcing material is nanomaterials such as carbon nanotubes or graphene. When a signal from a second sensor network indicating that the crack width or strength loss exceeds a preset threshold is received, the microcapsules rupture to release the repair agent, achieving self-repair.

[0301] The repair strategy optimization module is deployed on the cloud big data platform. It is used to calculate the repair efficiency based on the repair strength recovery data and generate the optimal repair strategy by combining repair cost, strength change, strain change and number of damage points using a multi-objective optimization algorithm.

[0302] The simulation and life assessment module, deployed on the cloud big data platform, is used to call the high-precision degradation prediction model and repair efficiency model, input real-time environmental factor data, material performance data and cumulative degradation history data, predict the material life and performance degradation trend, and evaluate the effect of each repair scheme on strength recovery and life extension.

[0303] The closed-loop update module is used to incrementally update the degradation coefficient and degradation index in the high-precision degradation prediction model by taking the strength recovery data, strain value and crack width after each repair as new samples, and driving the repair strategy optimization module to iteratively optimize until the health index is greater than or equal to the preset threshold and the remaining life is greater than or equal to the preset life value.

[0304] The API interface module is used to integrate the optimal repair strategy into external building management systems, construction management systems, or health monitoring systems, enabling intelligent, automated, and collaborative work across the entire project in material health monitoring, repair decision-making, and lifecycle management. The integration aims to achieve integrated intelligent collaboration across materials, structure, and operations and maintenance, with the following objectives:

[0305] (1) Improve the operation and maintenance efficiency of the entire project, reduce manual intervention, and improve the repair accuracy.

[0306] (2) Enable cross-platform collaboration so that different systems can share information and avoid data silos.

[0307] (3) Real-time feedback and decision-making: Adjust repair strategies and lifecycle management plans based on system feedback.

[0308] The specific process is as follows:

[0309] 1. Integration with Building Management System (BMS)

[0310] BMS System Overview: A Building Management System (BMS) is an intelligent system used to manage and control building facilities and equipment, encompassing multiple aspects such as energy management, environmental control, and equipment monitoring. In this step, a material health monitoring and repair feedback mechanism is integrated with the BMS to ensure real-time acquisition of material status data and to optimize repair and maintenance based on the material's health status.

[0311] 2. Integration methods and data sharing

[0312] Through Internet of Things (IoT) and cloud computing technologies, the BMS system will be able to receive, process, and analyze health data of the target structural materials transmitted from the first and second sensor networks. The specific integration steps are as follows:

[0313] (1) Sensor data acquisition: The health status of the target structural material (such as crack propagation, corrosion rate, stress change, etc.) is acquired in real time through the first sensor network and the second sensor network.

[0314] (2) Data transmission and storage: Sensor data from the first sensor network and the second sensor network are uploaded to the cloud big data platform via wireless communication protocols (such as LoRaWAN, NB-IoT, etc.) for centralized storage and management.

[0315] (3) BMS data interface: The BMS system obtains the health data of the target structural materials stored on the cloud big data platform through the API interface module and updates the status assessment of building facilities in real time.

[0316] (4) Decision support and feedback: Based on real-time monitoring data and material degradation prediction, the BMS system provides repair and maintenance suggestions for building facilities. For example, when the health index of a building structure is found to be lower than a preset threshold, the BMS will trigger a repair reminder.

[0317] 3. Automated Execution of Optimized Repair Strategies: After integration, the BMS system can not only achieve data sharing and real-time monitoring, but also automatically adjust the repair plan based on the health status of the target structural materials using the repair strategy optimization module in this embodiment. The repair process includes:

[0318] Repair plan generation: Based on the material's health status and degradation prediction model, the BMS system automatically generates the best repair plan, selecting appropriate repair materials, repair time, and repair methods.

[0319] Execution Feedback: After the repair is performed, the BMS system collects the repair effect data through the API interface module for subsequent evaluation and optimization.

[0320] 4. Integration with the construction management system:

[0321] Construction management systems are used to manage and monitor project progress, resource allocation, and quality control during the construction process. Integrating repair strategies and real-time monitoring systems will make material repair during construction more intelligent, ensuring material quality and safety.

[0322] The integration method and functional implementation are as follows:

[0323] By connecting to the construction management system via an API interface module, the health monitoring and repair strategies for the target structural materials are directly integrated with construction progress management and construction quality inspection. The specific steps are as follows:

[0324] (1) Real-time material status feedback: The health status of the material (such as crack length and corrosion degree) is fed back in real time through the first and second sensor networks, and the data is input into the construction management system.

[0325] (2) Repair decision and task allocation: The construction management system automatically assigns repair tasks to the construction team based on real-time data, ensuring that each construction worker is clear about the parts that need to be repaired, the repair materials to be used, and the repair time.

[0326] (3) Construction progress and quality control: The construction management system dynamically adjusts the construction progress in conjunction with the allocation of repair tasks and ensures that the repair process meets the quality standards.

[0327] (4) Data storage and feedback: Data during the repair process (such as the use of repair materials and performance evaluation after repair) will be recorded in real time to facilitate subsequent data analysis and life cycle management.

[0328] 5. Integration with health monitoring systems

[0329] The health monitoring system is mainly used to monitor the health status of structures and facilities in real time, such as monitoring cracks, deformation, and displacement in buildings, bridges, tunnels, etc. It integrates with the material health monitoring system of this embodiment (the material health monitoring system is a collective term for the first sensor network, the second sensor network, and the cloud big data platform) through an API interface module to achieve joint monitoring of the structural and material conditions.

[0330] Integration method and function implementation:

[0331] (1) Data sharing and collaborative work: The health monitoring system and the material health monitoring system share data through a unified data platform. In this platform, monitoring data from the structure (such as strain, displacement, vibration, etc.) and material health data (such as crack propagation, corrosion degree, etc.) are cross-analyzed to provide a comprehensive health assessment.

[0332] (2) Multi-dimensional monitoring and diagnosis: After integration, the system can diagnose potential structural risks through multi-dimensional data analysis and provide repair decision support based on the health status of materials and structures.

[0333] (3) Joint repair strategy: Through data analysis, the system can propose a global repair plan that combines structural repair and material repair to improve repair efficiency and cost-effectiveness.

[0334] 6. Advantages of system integration

[0335] (1) Intelligent management: The integrated system can realize automated decision-making and real-time repair, which greatly improves the operation and maintenance efficiency and repair effect of engineering projects.

[0336] (2) Cross-platform collaboration: Through a unified data platform, different systems (such as BMS, construction management system, and health monitoring system) can share data, avoid information silos, and achieve seamless collaboration.

[0337] (3) Comprehensive assessment and prediction: The integrated system can comprehensively assess and predict the life of materials and structures based on real-time and historical data, ensuring long-term safety and reliability.

Claims

1. A multi-environment coupled degradation prediction and adaptive repair synergistic optimization method, characterized in that, Includes the following steps: Step 1: Collect environmental factor data and material performance data by deploying a first sensor network on the surface of the target structure, and construct a degradation prediction model that includes the degradation coefficient and degradation index to be optimized for materials coupled with multiple environmental loads; the environmental factor data includes temperature, humidity, salinity and pressure, and the material performance data includes initial compressive strength, elastic modulus, coefficient of thermal expansion and porosity. Step 2: Using the environmental factor data and material performance data collected in Step 1 as the original samples, and combining them with the spatiotemporal degradation history data corresponding to the original samples, a standardized training dataset is formed after missing value processing and standardization. The degradation coefficient and degradation index to be optimized in the degradation prediction model of Step 1 are then globally optimized using the quantum annealing algorithm. The optimized degradation coefficient and degradation index are then fixed, and an artificial neural network is used to establish a nonlinear mapping of environmental performance to obtain a high-precision degradation prediction model. The degradation history data includes the degradation rate, crack propagation, and strength loss of materials under different environments. Step 3: Embed an adaptive repair material and deploy a second sensor network in the target structure. The adaptive repair material includes a self-healing polymer matrix, a repair agent, and a reinforcing material. The self-healing polymer matrix is ​​epoxy resin or polyurethane resin, the repair agent is embedded in the self-healing polymer matrix in the form of microcapsules or fiber mesh, and the reinforcing material is carbon nanotubes or graphene. The second sensor network is integrated with the adaptive repair material and real-time data on crack width, strain value, and corrosion rate are collected as health data. When the second sensor network detects cracks or strength loss exceeding a threshold, the microcapsules rupture and release the repair agent. Simultaneously, post-repair strength recovery data is continuously collected and transmitted back to the cloud big data platform. The repair efficiency is calculated based on the post-repair strength recovery data, and an initial repair strategy is generated using an optimization objective function, taking into account repair cost, strength change, strain change, and number of damage points. Step 4: Write the high-precision degradation prediction model from Step 2 and the repair efficiency from Step 3 into the simulation platform. Input real-time environmental factor data, material performance data, and historical degradation data accumulated from Step 3 to predict the material life and performance degradation trend. Based on the initial repair strategy generated in Step 3, evaluate the effect of each repair scheme on material strength recovery and life extension to obtain life prediction results and repair effect evaluation results. Use optimization algorithms to perform multi-objective optimization on repair time and repair material type to form the optimal repair strategy. Step 5: Perform repair on the target structure according to the optimal repair strategy obtained in Step 4, and feed the repair effect back to the high-precision degradation prediction model in Step 2 for dynamic updating of model parameters. With repair cost, strength change, strain change, and number of damage points as optimization objectives, use the same multi-objective optimization algorithm as in Step 4 to iteratively optimize the repair time, repair materials, and repair methods to obtain a dynamically updated repair strategy. Calculate the health index based on the material compressive strength at time t and the initial compressive strength output by the high-precision degradation prediction model in Step 2, and calculate the remaining life based on the health index and the degradation coefficient obtained in Step 2. Step 6, closed-loop optimization and repair feedback update based on big data, includes the following sub-steps: Step 6.1: Upload the high-precision degradation prediction model obtained in Step 2, the repair efficiency obtained in Step 3, and the optimal repair strategy obtained in Step 4 to the cloud big data platform. Step 6.2: Continuously acquire environmental factor data, material performance data, crack width, strain value, corrosion rate, and post-repair strength recovery data through the first sensor network in Step 1 and the second sensor network in Step 3. After data cleaning, missing value filling, and normalization, new samples are formed. Step 6.3: Use the new samples from Step 6.2 to incrementally update the degradation coefficient and degradation index in Step 2, and then send the updated degradation coefficient and degradation index back to the high-precision degradation prediction model in Step 2 to replace the original fixed values ​​and dynamically refresh the model parameters. Step 6.4: Input the high-precision degradation prediction model refreshed in Step 6.3 back into the simulation platform in Step 4. Combine it with the strength recovery data accumulated in Step 6.2 after repair, recalculate the repair efficiency, and use the same multi-objective optimization algorithm as in Step 4 to iteratively optimize the repair time and the type of repair material. The optimization result is used as the new current optimal repair strategy. Step 6.5: After performing the repair according to the new current optimal repair strategy in Step 6.4, continue to collect the strength recovery data, strain value and crack width after repair through the second sensor network. Repeat steps 6.2 to 6.4 with the newly added data to obtain the updated current optimal repair strategy. Step 6.6: Repeat steps 6.2 to 6.5 to make the high-precision degradation prediction model in step 2 converge continuously with the measured data until the health index output in step 2 is greater than or equal to the preset threshold and the remaining lifespan is greater than or equal to the preset lifespan value.

2. The method according to claim 1, characterized in that, The formula for constructing the environmental load-material coupled degradation model described in step 1 is as follows: （1）， In the formula: express The compressive strength of the material at all times; Indicates the initial compressive strength; This function represents the influence of environmental factors on material properties, specifically the effects of temperature, humidity, salinity, and pressure. It is fitted to environmental data to describe the degradation effects of these factors on the material. The degradation coefficient, expressed in units of 1 / year, describes the rate of material degradation under harsh environments and depends on environmental factors and the type of material; the degradation rate under different environments is obtained through experiments. Indicates time, in years, indicating the service life of the material; The degradation index represents the degree of acceleration of material degradation; this degradation index is obtained by fitting experimental data under different material and environmental conditions. The values ​​are different.

3. The method according to claim 1, characterized in that, The formula for the quantum degeneration algorithm described in step 2 is as follows: （4）， In the formula: Corresponding energy state The probability of; It represents the energy state, specifically the optimization target of the degradation coefficient and degradation index; Represents the Boltzmann constant; This represents the temperature change during quantum annealing; represents the partition function, used to normalize the probability distribution; e represents the base of the natural logarithm, which is 2.71828, used for exponential calculation; The artificial neural network is a deep neural network or a convolutional neural network. The deep neural network is used to perform regression prediction on the compressive strength and degradation rate of materials, using the mean squared error as the loss function. The formula for the mean squared error is as follows: （5）， In the formula: Represents the loss function; Indicates the true value (the degree of material degradation); Indicates the model's predicted value; Indicates the number of samples; Convolutional neural networks are used to extract features and identify defects from images or spatial distribution data of cracks on material surfaces.

4. The method according to claim 1, characterized in that, The formula for the repair efficiency described in step 3 is as follows: （8）， In the formula: express Real-time repair efficiency, in percentages (%) This represents the initial repair efficiency, which is usually set to 1; This represents the repair rate coefficient, measured in units of 1 / year, and is related to environmental factors and material properties. Indicates the repair time, in years; The optimization objective function is: （9）， In the formula: The objective function for the repair decision represents the overall objective to be optimized during the repair process, which is usually achieved by taking multiple factors into account through weighted summation. This represents the total number of damage types; Indicates the first Weighting coefficients for different damage types; Indicates the first Changes in compressive strength for different damage types; Indicates the first Strain variation weighting coefficients for different damage types; Indicates the first Strain changes for each damage type, expressed in % . Indicates the first Weighting coefficients for repair costs of different damage types; Indicates the first Repair costs for different types of damage.

5. The method according to claim 1, characterized in that, The optimization algorithm described in step 5 is either particle swarm optimization or genetic algorithm; The formula for the particle swarm optimization algorithm is as follows: （10）， （11）， In the formula: Indicates the first The particle in the first The speed of each iteration; Indicates the first The particle in the first The position of the next iteration; Indicates inertial weight, which controls the inertia of particles; , It represents the acceleration constant, which controls the tendency of particles towards local and global optima; , Represents a random number, used to introduce randomness; Indicates the first The historical optimal solution for each particle; This represents the global optimal solution for all particles; Indicates the first The velocity of a particle in the kth iteration; Indicates the first The position of the particle in the kth iteration; The genetic algorithm includes the following steps: Step 5.3.1, Selection Operation: Select individuals with higher fitness as parents, with selection probability. It is directly proportional to the individual's fitness value, as shown in the following formula: （12）， In the formula: Indicates the first The probability of an individual's choice; Indicates the first The fitness value of an individual is usually the value of the objective function; This indicates the population size, and the total number of individuals in the population. Step 5.3.2, Crossover Operation: Parent individuals generate new offspring individuals through a crossover operation. The crossover individuals... Represented as: （13）， In the formula: This represents the offspring individuals generated after the crossover operation; and Indicates the selected parent individual; Step 5.3.3, Mutation Operation: Mutation is performed on the offspring individuals after crossover to increase the diversity of solutions, as shown in the following formula: （14）， In the formula: Indicates the mutated individual; This represents the offspring individuals generated after the crossover operation; Step 5.3.4, Fitness Function: To evaluate the merits of each individual (repair strategy), a fitness function is defined. The goal is to minimize repair time and cost while maximizing repair efficiency. （15）， In the formula: Represents an individual fitness value; , , This represents the weight coefficient of each optimization objective in the repair strategy, used to balance the influence of different objectives; Repair time indicates the time required to perform the repair. Repair cost refers to the expense required for repair; repair efficiency refers to the efficiency with which the material strength is restored after repair. Step 5.3.5: Update Operation: After each generation iteration, the population is updated through a replacement operation, selecting individuals with higher fitness to ensure the population evolves towards the optimal solution. （16）， In the formula: Indicates the updated population; This refers to a subset of individuals from an old population that have undergone selection, crossover, and mutation. It represents the new individual after the mutation.

6. The method according to claim 1, characterized in that, The formula for calculating the health index mentioned in step 5 is as follows: （17）， In the formula: for The health index of a material at any given time represents the ratio of its current strength to its initial strength. for The compressive strength of the material at any given time is obtained from the degradation prediction model; Indicates the initial compressive strength of the material; The formula for calculating the remaining lifetime is as follows: （18）， In the formula: Indicates remaining life expectancy, in years; The degradation coefficient describes the rate of material degradation under specific conditions.

7. A multi-environment coupled degradation prediction and adaptive repair collaborative optimization system, characterized in that, include: The first sensor network is deployed on the surface of the target structure to collect environmental factor data and material performance data, and output them to the degradation prediction modeling module. The degradation prediction modeling module is connected to the first sensor network and is used to construct a degradation prediction model containing degradation coefficients and degradation indices to be optimized based on the environmental factor data and material performance data. The degradation coefficients and degradation indices are globally optimized using the quantum annealing algorithm, and an environmental performance nonlinear mapping is established using an artificial neural network to obtain a high-precision degradation prediction model. A cloud-based big data platform is used to store the high-precision degradation prediction model, the repair efficiency model, and the optimal repair strategy. The second sensor network, integrated with the adaptive repair material, is embedded inside the target structure to collect data on crack width, strain value, corrosion rate, and post-repair strength recovery in real time, and uploads it to the cloud-based big data platform. An adaptive repair material is used to release a repair agent by rupturing microcapsules when a signal from a second sensor network indicates that the crack width or strength loss exceeds a preset threshold, thereby achieving self-repair. The repair strategy optimization module is deployed on the cloud big data platform. It is used to calculate the repair efficiency based on the repair strength recovery data and generate the optimal repair strategy by combining repair cost, strength change, strain change and number of damage points using a multi-objective optimization algorithm. The simulation and life assessment module, deployed on the cloud big data platform, is used to call the high-precision degradation prediction model and repair efficiency model, input real-time environmental factor data, material performance data and cumulative degradation history data, predict the material life and performance degradation trend, and evaluate the effect of each repair scheme on strength recovery and life extension. The closed-loop update module is used to incrementally update the degradation coefficient and degradation index in the high-precision degradation prediction model by taking the strength recovery data, strain value and crack width after each repair as new samples, and driving the repair strategy optimization module to iteratively optimize until the health index is greater than or equal to the preset threshold and the remaining life is greater than or equal to the preset life value.

8. The system according to claim 7, characterized in that, It also includes an API interface module for outputting the optimal repair strategy to an external building management system, construction management system, or health monitoring system.

9. The system according to claim 7, characterized in that, The adaptive repair material includes a self-healing polymer matrix, a repair agent, and a reinforcing material. The self-healing polymer matrix is ​​epoxy resin or polyurethane resin, the repair agent is embedded in the self-healing polymer matrix in the form of microcapsules or fiber mesh, and the reinforcing material is nanomaterials such as carbon nanotubes or graphene.

10. A computer program product, characterized in that, The product includes a storage medium and computer-readable instructions stored on the medium, which, when executed by a computer, implement the method of any one of claims 1 to 6.

Images