Concrete degradation simulation method and system under continuous load-chloride ion coupling

Abstract

The invention relates to a concrete degradation simulation method and system under continuous load-chloride ion coupling, and solves the problems that systematic research on overall mechanical property degradation of a concrete structure under the load and chloride ion coupling effect is lacked, and the degradation condition of the concrete structure in an actual service environment is difficult to comprehensively reflect. The method comprises the following steps: applying a continuous load to a test block, preparing a solution according to a stress type to determine a chloride ion diffusion direction, applying an electric field to accelerate migration and maintain test stability, and after an experiment, measuring the penetration depth and calculating a diffusion coefficient; in the process, multi-dimensional data such as cracks and acoustic emission are collected and subjected to PCA dimension reduction and redundancy removal, then damage factors are inversed through deep learning modeling in combination with dual-stage optimization and physical constraint, and finally damage and chloride ion parameters are fused to form gradient parameters fitting actual degradation. The method has the following effects that the concrete degradation process is accurately simulated, a basis is provided for durability evaluation and life prediction, and the engineering safety and economy are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building material durability evaluation, and in particular to a concrete deterioration simulation method and system under sustained load-chloride ion coupling. BACKGROUND

[0002] The coastal service environment has characteristics such as high concentration of erosive ions and accelerated erosion under sustained load on concrete structures, which seriously threatens the durability and safe application of concrete. Therefore, the research and accurate evaluation of the chloride ion penetration resistance of concrete under load conditions are particularly important for the durability design and life prediction of concrete.

[0003] At present, the methods for testing and evaluating the chloride ion penetration resistance of concrete mainly include natural diffusion method, dry-wet cycle acceleration method and external electric field acceleration diffusion method. Among them, the external electric field acceleration diffusion method can obtain experimental results in a shorter time by applying an external electric field to accelerate the migration of chloride ions. In addition, some studies establish a structural mechanics model by using finite element analysis software, assign the deteriorated material parameters to the model, and then analyze the overall mechanical performance degradation of the structure. These methods provide technical support for the durability research of concrete to some extent.

[0004] The prior art has the following disadvantages: first, the existing methods are mostly based on single factor (such as load or chloride ion erosion) research, and lack of systematic research on the overall mechanical performance degradation of concrete structures under the coupling action of load and chloride ion, which is difficult to fully reflect the deterioration of concrete structures in actual service environment. Secondly, the existing experimental device lacks accurate determination of the chloride ion diffusion coefficient under sustained load, and cannot clearly determine the influence mechanism of load on chloride ion erosion. SUMMARY

[0005] In order to accurately simulate the deterioration process of concrete, provide scientific basis for durability evaluation and life prediction, and improve engineering safety and economy, the present application provides a concrete deterioration simulation method and system under sustained load-chloride ion coupling.

[0006] In the first aspect, the present application provides a concrete deterioration simulation method under sustained load-chloride ion coupling, which adopts the following technical scheme:

[0007] A concrete deterioration simulation method under sustained load-chloride ion coupling, comprising:

[0008] The concrete test block is prepared by using a preset preparation method, after standard curing and saturated solution pretreatment, the support points are adjusted according to the three-point bending specification, and the test block is placed, the tensile zone and the compression zone of the test block are determined, the test block is loaded by using a preset loading device, the load is continuously increased until the peak value is reached, the peak value is recorded as the ultimate load of the test block by using a preset sensor and a load digital display panel, and a specific proportion of the continuous tensile / compression load is applied and maintained based on the ultimate load, and the sensor and the load digital display panel are used for real-time monitoring and feedback control, so that the stability of the load value is ensured, wherein the tensile load corresponds to the tensile zone of the test block, and the compression load corresponds to the compression zone of the test block;

[0009] The test components are assembled based on the stress area of the test block, the solution is adjusted according to the stress type to determine the diffusion direction of chloride ions, the electrodes are connected, and an external electric field is applied to accelerate the migration of chloride ions; during this period, the electrode solution is supplemented in a preset solution supplementing method to maintain the stability of the test environment, and the entire chloride ion migration stage continues until the preset experiment ending requirement is met;

[0010] After the experiment is completed, the power supply is disconnected, the test block is processed and sprayed with a preset color developing agent, the chloride ion penetration depth is measured, and the chloride ion diffusion coefficient under different loading degrees and different stress types is calculated respectively;

[0011] The diffusion coefficient data under different conditions are combined to establish the correlation between the loading degree, the stress type and the chloride ion diffusion coefficient.

[0012] By using the above technical scheme, the method simulates the coupling effect of load and chloride ions, accurately controls the experimental conditions, accelerates the migration of chloride ions, accurately measures the chloride ion penetration depth and calculates the diffusion coefficient, and establishes the correlation. It can provide reliable data for the durability research of concrete, help to optimize the design of concrete structure, prolong the service life, and has important engineering application value.

[0013] In a second aspect, the present application provides a concrete deterioration simulation system under continuous load-chloride ion coupling, which adopts the following technical scheme:

[0014] A concrete deterioration simulation system under continuous load-chloride ion coupling, comprising a memory, a processor and a program stored on the memory and executable on the processor, which can be loaded and executed by the processor to realize the concrete deterioration simulation method under continuous load-chloride ion coupling as described in the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 It is a whole flow schematic diagram of a concrete deterioration simulation method under continuous load-chloride ion coupling according to an embodiment of the present application.

[0016] Figure 2is a structural schematic diagram of a loading device configured with a chloride ion test component in an embodiment of the present application.

[0017] Figure 3 is a structural schematic diagram of a chloride ion test component in an embodiment of the present application.

[0018] Figure 4 is a top view of a chloride ion test component in an embodiment of the present application.

[0019] Figure 5 is a partial schematic diagram of a chloride ion test component in an embodiment of the present application, showing the assembly relationship of a sponge, a titanium mesh, a solution tank, and a test block. DETAILED DESCRIPTION

[0020] The present application is further described in detail below with reference to the accompanying drawings.

[0021] Reference Figure 1 A sustained load-chloride ion coupling concrete deterioration simulation method disclosed in the present application comprises:

[0022] In step S100, a concrete test block is prepared by a preset preparation method. After standard curing and saturated solution pretreatment, the test block is placed by adjusting the support points according to the three-point bending specification, the tensile zone and the compression zone of the test block are determined, the test block is loaded by a preset loading device, the load is continuously increased until the peak value is reached, the peak value is recorded as the ultimate load of the test block by a preset sensor and a load digital display panel, a specific proportion of sustained tensile / compression load is applied and maintained based on the ultimate load, and real-time monitoring and feedback control are performed by the sensor and the load digital display panel to ensure the stability of the load value, wherein the tensile load corresponds to the tensile zone of the test block, and the compression load corresponds to the compression zone of the test block. The specific loading device can be seen in Figure 2 .

[0023] The pre-set preparation method: according to the experimental design requirements, the pre-determined concrete proportioning and mixing process, to ensure that the concrete test block has specific performance. Standard curing: the concrete test block is placed in a standard curing chamber, the temperature is controlled at 20±2℃, the relative humidity is above 95%, and it is cured for a certain period of time to ensure that the concrete reaches the design strength. Saturated solution pretreatment: the cured test block is immersed in saturated Ca(OH)2solution to enhance its erosion resistance. Three-point bending specification: a mechanical test method that tests the bending performance of the test block by applying concentrated load on it. Loading device: a device for applying load to the concrete test block, usually including loading wheels, bending clamps, etc. Sensor: a device for monitoring the load on the test block in real time, which can convert the load signal into an electrical signal output. Load digital display panel: a device connected with the sensor, used to display the real-time load value for the convenience of observation and recording by the experimenters. Continuous tension / compression load: during the experiment, a certain amount of tensile or compressive load is applied to the test block and maintained, simulating the stress state of the concrete structure in actual engineering. Tension zone / compression zone: the upper and lower sides of the concrete test block respectively bear tensile stress and compressive stress under the action of load.

[0024] The overall process is as follows:

[0025] First, accurately weigh the cementitious materials, sand and stones according to the designed proportioning, and put these materials into the mixer for low-speed pre-mixing for 20-30 seconds. Then add water and admixtures in the predetermined proportion, and continue to mix at high speed for 3-5 minutes to ensure uniform mixing of the concrete. After mixing is completed, pour the fresh concrete into the corresponding size mold quickly, and make the structure compact through vibration equipment to ensure the internal quality of the concrete. After pouring is completed, carry out film coating treatment on the test block, and remove the mold after 24 hours of standing. The test block after demolding needs to be immediately transferred to the standard curing chamber for curing under the conditions of temperature 20±2℃ and relative humidity above 95% until the specified age (3-90 days) to ensure that the concrete reaches the design strength. After the curing age arrives, immerse the test block in saturated Ca(OH)2solution for 3 days for pretreatment to enhance its erosion resistance. After pretreatment is completed, place the soaked concrete beam on the support points according to the three-point bending test specification, wipe the surface dry, and seal the front and back surfaces with resin to determine the tension zone (lower side) and compression zone (upper side) of the concrete under continuous load. Finally, directly load the test block in contact with the bending clamp through the loading wheel disc of the loading device, monitor the load change in real time using the sensor, and display the load value in real time through the load digital display panel. The above process can be referred to Figure 2, the sensor monitors the load on the test block in real time and converts it into an electrical signal, and the load digital display panel displays the real-time load value synchronously; when the load increases to the "peak value" (i.e. the maximum load value at which the load no longer rises and the test block is about to be damaged), the peak value is recorded, which is the ultimate load of the test block. After obtaining the ultimate load of the test block, the required sustained tensile / compressive load is accurately controlled and monitored in real time to ensure the stability of the load during the experiment. Specifically, the load digital display panel compares the "current load value" monitored in real time with the "target load value" preset, and if there is a deviation (such as the current load being lower than the target value, which may be due to insufficient loading force caused by slight deformation of the test block; or higher than the target value, which may be due to overloading caused by the inertia of the loading device), an "adjustment signal" is immediately sent to the control module of the loading device; after receiving the adjustment signal, the loading device drives the execution component (such as the loading wheel disc, hydraulic assembly) to make fine adjustments: if the load is low, increase the loading force (such as pushing the loading wheel disc to apply more pressure to the test block); if the load is high, reduce the loading force (such as slowing down the loading wheel disc or slightly retreating). The specific ratio needs to be set according to the experimental target (such as simulating the actual service load level of concrete structures), which is an adjustable parameter in the experimental design stage, and the core principle is to take the "ultimate load of the test block" as the benchmark (such as taking 50%, 60%, 70% of the ultimate load, etc.), and the common value range is usually 50%-80% of the ultimate load.

[0026] Step S200, based on the stress region of the test block, assemble the test component, adjust the solution according to the stress type to determine the diffusion direction of chloride ions, connect the electrode and apply an external electric field to accelerate the migration of chloride ions; during this period, the electrode solution is supplemented in a timely manner according to the preset solution supplementing method to maintain the stability of the test environment, and the whole chloride ion migration stage continues until the preset experimental end requirement is met.

[0027] Among them, the test component: auxiliary device for experiment, including upper and lower solution tank, titanium mesh, sponge tank, etc., used to simulate the actual service environment and accelerate the migration of chloride ions. Chloride ion diffusion direction: the migration path of chloride ions in the concrete test block, determined by the arrangement of different solutions in the solution tank. External electric field: electric field applied by electrode, used to accelerate the migration of chloride ions in concrete. Preset solution supplementing method: according to the experimental requirements, regularly supplement the electrode solution to maintain the stability of the test environment. The preset experimental end requirement (all need to be set in advance), the core is as follows: 1, reach the preset migration time (such as "7 days of migration" in the file example); 2, the penetration depth of chloride ions is enough to measure the threshold value (to avoid too shallow to cause large measurement error); 3, the chloride ions form a clear concentration gradient in the test block (to meet the subsequent diffusion coefficient calculation).

[0028] The overall process (the part involving the test component can refer to Figures 3 to 5 ) is as follows:

[0029] 1. Test Component Assembly: Two pieces of appropriately sized titanium mesh are placed on the top of the loaded specimen and connected by copper wire, whose length is adjustable to avoid the flexural jig. A prefabricated upper solution tank (made of acrylic) is then bonded to the top of the specimen using glass adhesive. Finally, the specimen is wrapped on three sides with sponge, which is filled with electrode solution and secured with a prefabricated, foldable sponge tank (made of acrylic). The nuts on the sponge tank are adjustable in tension, and the open bottom of the sponge is in direct contact with the solution in the lower solution tank. A piece of titanium mesh is placed between the lower surface of the specimen and the sponge. Before the experiment, the titanium mesh on the upper and lower surfaces are connected to the corresponding electrodes.

[0030] 2. Chloride ion diffusion direction adjustment: Adjust the solutions in the upper and lower solution tanks according to the load type to determine the chloride ion diffusion direction. For tensile loads, the upper solution tank contains NaOH solution and the lower contains NaCl solution. The chloride ion diffusion direction is from the tensile side to the compressive side. For compressive loads, the upper solution tank contains NaCl solution and the lower contains NaOH solution. The chloride ion diffusion direction is from the compressive side to the tensile side.

[0031] 3. Apply an external electric field to accelerate the migration of chloride ions: After the selected solution is injected into the upper and lower solution tanks, the power supply electrodes are connected to the corresponding upper and lower titanium meshes and adjusted to the appropriate voltage or current to accelerate the migration of chloride ions. Figures 3 to 5 The electrodes A and B shown are made of titanium mesh (for excellent conductivity and resistance to chloride ion corrosion) and connected to an external power supply via copper wire (away from the flexural clamp to prevent short circuits). Electrode A corresponds to the tensile zone of the concrete specimen (e.g., the lower solution tank, filled with NaCl solution), while electrode B corresponds to the compressive zone (e.g., the upper solution tank, filled with NaOH solution). These electrodes are arranged along the load direction (tension / compression), creating an electric field gradient. Electrode A is connected to the anode (positive potential) of the power supply, and electrode B to the cathode (negative potential), forming an electric field within the specimen from the tensile zone (A) to the compressive zone (B).

[0032] 4. Solution replenishment: During the experiment, the electrode solution should be replenished regularly according to the preset method to ensure the stability of the solution concentration and volume and maintain the stability of the test environment.

[0033] For example, assume the specimen is in tension during the experiment. A 0.1 mol / L NaOH solution is added to the upper solution tank, while a 0.1 mol / L NaCl solution is added to the lower solution tank. Titanium mesh is placed in each tank, connected to power electrodes, and a 10V DC voltage is applied to accelerate the migration of chloride ions toward the compressive side. During the experiment, the electrode solution is replenished every two hours to ensure stable solution concentration and volume. This method effectively simulates the migration of chloride ions in concrete under actual service conditions and provides accurate data for subsequent diffusion coefficient calculations.

[0034] Step S300, after the experiment, disconnect the power, handle the test block and spray the preset color developing agent, measure the chloride ion penetration depth, calculate the chloride ion diffusion coefficient under different loading degrees and different stress types according to the experimental parameters.

[0035] Where, the chloride ion penetration depth: the depth of chloride ion diffusion in the concrete test block, measured after color development by color developing agent. Chloride ion diffusion coefficient: a parameter describing the diffusion rate of chloride ion in concrete, calculated from experimental data. Preset color developing agent: a chemical reagent used to indicate the diffusion depth of chloride ion, usually AgNO3 solution. Experimental parameters: parameters recorded during the experiment, such as voltage, temperature, load, etc., used to calculate the chloride ion diffusion coefficient.

[0036] The overall process is as follows:

[0037] 1. Post-experiment processing: After the experiment, immediately disconnect the power and remove the test block from the loading device. Rinse the test block thoroughly and dry it, ensuring that the surface is free of residual solution. Use a cutting machine to cut the test block perpendicular to the direction of erosion ion diffusion, ensuring that the cut surface is flat.

[0038] 2. Chloride ion penetration depth measurement: Spray the preset color developing agent (such as 0.1 mol / L AgNO3 solution) on the cut surface of the test block. Chloride ions react with AgNO3 to form insoluble AgCl precipitate, forming a white color development area. By measuring the boundary of the color development area, the chloride ion penetration depth is determined.

[0039] 3. Chloride ion diffusion coefficient calculation: a. According to the experimental parameters (such as voltage, temperature, load, etc.) and the chloride ion penetration depth, the chloride ion diffusion coefficient is calculated according to Fick's second law. b. The formula of Fick's second law is: where, is the chloride ion concentration at a depth of x from the concrete surface and a diffusion time of t, is the initial concentration, D is the diffusion coefficient, t is the time, and erfc is the complementary error function. c. Through the recorded chloride ion penetration depth and time in the experiment, the diffusion coefficient D is back calculated.

[0040] Example: Suppose in the experiment, the test block is under the action of sustained tensile load, 10V voltage is applied to accelerate the migration of chloride ions, and the experimental time is 7 days. After the experiment, the test block is cut and sprayed with 0.1 mol / L AgNO3 solution, and the chloride ion penetration depth is measured to be 10 mm. According to the experimental parameters (voltage 10V, temperature 20℃, time 7 days), combined with Fick's second law, the chloride ion diffusion coefficient D is calculated to be 1.5×10−12 m² / s. In this way, the chloride ion diffusion performance of concrete under different loading degrees and stress types can be accurately evaluated.

[0041] Step S400, combine the diffusion coefficient data under different conditions to establish the correlation between the loading degree, stress type and chloride ion diffusion coefficient.

[0042] where loading degree: the size of the load applied on the concrete test block, usually expressed in terms of stress or strain. Stress type: the type of load the test block is subjected to, such as tensile load or compressive load. Chloride ion diffusion coefficient: a parameter describing the rate of chloride ion diffusion in concrete, calculated through experiments. Correlation model: a mathematical model used to describe the relationship between loading degree, stress type and chloride ion diffusion coefficient.

[0043] The overall process is as follows: 1. Data organization: collect chloride ion diffusion coefficient data under different loading degrees and stress types, including the load size, stress type (tension or compression) recorded in the experiment and the corresponding chloride ion diffusion coefficient. 2. Establish correlation model: use regression analysis or machine learning algorithms to establish a relationship model between loading degree, stress type and chloride ion diffusion coefficient. For example, linear regression model or nonlinear regression model can be used to fit the data. Assuming that the relationship between diffusion coefficient D and loading degree (expressed as stress σ) and stress type (tension or compression) can be expressed as: where a, b and c are model parameters, and type is an indicator variable representing stress type (tension or compression). 3. Model verification and optimization: use part of the experimental data to train the model, then use the remaining data to verify, ensure the accuracy and reliability of the model. If the model prediction error is large, you can adjust the model parameters or try other more complex models, such as polynomial regression or neural network model. 4. Results application: apply the established model to practical engineering, according to the specific loading conditions and stress type, predict the diffusion coefficient of chloride ion in concrete structure, so as to evaluate the durability of concrete structure.

[0044] where assembling test components based on the stress area of the test block includes:

[0045] Step S201, configure the chloride ion test component containing solution tank, titanium mesh, conductive connecting piece and fixing device, and clean and pretreat the surface of the tensile and compressive regions of the test block. The chloride ion test component can refer to the one shown in Figures 3 to 5 .

[0046] Solution tank: A container used to hold the electrolyte solution (such as sodium chloride solution), usually made of insulating materials (such as acrylic plate), used to simulate the actual service environment of chloride ion corrosion conditions. Titanium mesh: A metal mesh with good electrical conductivity, used as an electrode to accelerate the migration of chloride ions in concrete. Conductive connector: A conductive component used to connect the titanium mesh and the power supply, such as copper wire, to ensure stable current through the titanium mesh. Fixing device: A device used to fix the solution tank on the concrete test block, usually made of glass glue, nuts and other materials, to ensure that the solution tank does not shift during the experiment. Cleaning and pretreatment: Clean the surface of the test block in the tension area and the compression area, remove surface impurities and dust, to ensure good contact between the titanium mesh and the test block surface.

[0047] The overall process is as follows: 1. Configure the chloride ion test components: a. Solution tank: Pre-fabricate the upper and lower solution tanks according to the size of the test block, usually made of acrylic plate, to ensure that the size is suitable for the test block. b. Titanium mesh: Select the appropriate size of titanium mesh for the tension area and compression area of the test block. The size of the titanium mesh should be slightly smaller than the inner diameter of the solution tank to ensure that it can be completely immersed in the solution. c. Conductive connector: Prepare copper wire and other conductive connectors to connect the titanium mesh and the power supply to ensure stable current flow. d. Fixing device: Prepare glass glue, nuts and other fixing devices to fix the solution tank on the test block to ensure that it is stable during the experiment. 2. Test block surface cleaning and pretreatment: Use sandpaper or steel wire brush to gently polish the surface of the test block in the tension area and the compression area to remove surface dust and impurities. Wipe the test block surface with a clean damp cloth to ensure that the surface is clean and dust-free. Dry the test block surface to ensure that there is no water residue on the surface to ensure good contact between the titanium mesh and the test block surface.

[0048] Step S202, based on the pretreated test block tension area and compression area position, the titanium mesh is correspondingly laid on the stress surface and connected by the conductive connector to realize stable conductive connection, and the solution tank is positioned and assembled outside the test block stress area by the fixing device to form a test assembly suitable for the stress area.

[0049] The overall process is as follows: 1. Titanium mesh laying and conductive connection: a. According to the position of the tensile and compressive regions of the test block, lay the titanium mesh with appropriate size on the upper and lower surfaces of the test block. b. Use conductive connectors (such as copper wire) to connect the titanium mesh with the power supply, ensuring that the current can pass stably. The length of the copper wire can be adjusted to avoid the anti-bending clamp and avoid interfering with the experimental device. 2. Solution tank positioning and fixing: a. The pre-prepared solution tank (made of acrylic plate) is bonded to the outside of the stress area of the test block through the fixing device (such as glass glue). b. Ensure that the solution tank is tightly attached to the surface of the test block without air bubbles and gaps to ensure that the solution can be evenly distributed and effectively contact the surface of the test block. 3. Assembly of the overall test assembly: a. Check if the connection between the titanium mesh and the solution tank is firm to ensure that it will not loosen during the experiment. b. Confirm that the conductive connector is connected correctly without short circuit or open circuit phenomenon. c. Finally, form a test assembly that fits the stress area, preparing for the subsequent chloride ion diffusion experiment.

[0050] Adjusting the solution according to the stress type to determine the chloride ion diffusion direction includes:

[0051] Step S20A, determine the type of sustained load that the test block is subjected to, and mark the specific spatial position of the tensile and compressive regions of the test block. The type of sustained load includes tensile load and compressive load.

[0052] Among them, the type of sustained load refers to the nature of the load that the test block is subjected to during the experiment, which is divided into tensile load (tensile) and compressive load (compressive). Tensile region: the region of the test block that is subjected to tensile stress under the action of the load. Compressive region: the region of the test block that is subjected to compressive stress under the action of the load. Mark: mark the specific position of the tensile and compressive regions on the surface of the test block by physical or chemical methods for subsequent operation.

[0053] The overall process is as follows:

[0054] 1. Determine the load type: according to the experimental design, determine the type of sustained load that the test block will be subjected to, i.e. tensile load or compressive load. For example, if the experiment simulates the load-ion coupling effect of the tensile region of the arch waist of the tunnel concrete segment structure, the test block is subjected to tensile load; if it simulates the compressive region outside the top and bottom of the tunnel, the test block is subjected to compressive load.

[0055] 2. Mark the stress area: mark the specific position of the tensile and compressive regions on the surface of the test block. Physical markers such as markers or stickers can be used, or chemical markers such as a small amount of distinguishable chemical reagent can be used. Ensure that the mark is clear, accurate and can clearly distinguish between the tensile and compressive regions.

[0056] Step S20B, based on the marked stress area, configure the solution according to the load type, as follows: under tensile load, the upper solution tank corresponding to the compression zone is injected with NaOH solution, and the lower solution tank corresponding to the tension zone is injected with NaCl solution; under compression load, the upper solution tank corresponding to the compression zone is injected with NaCl solution, and the lower solution tank corresponding to the tension zone is injected with NaOH solution.

[0057] Wherein, solution tank: a container for holding electrolyte solution (such as NaCl solution or NaOH solution), usually made of insulating material (such as acrylic plate). NaCl solution: sodium chloride solution, as a source of chloride ions, used to simulate the chloride ion corrosion conditions in the actual service environment. NaOH solution: sodium hydroxide solution, used to form an electrochemical gradient in the solution tank, promoting the migration of chloride ions. Chloride ion diffusion direction: the migration path of chloride ions in the concrete test block, determined by the arrangement of different solutions in the solution tank.

[0058] The overall process is as follows:

[0059] 1. Configure the solution according to the load type: a. Tensile load: When the test block is under tensile load, the upper solution tank corresponding to the compression zone is injected with NaOH solution, and the lower solution tank corresponding to the tension zone is injected with NaCl solution. This setting is to simulate the migration of chloride ions from the tension zone to the compression zone. b. Compression load: When the test block is under compression load, the upper solution tank corresponding to the compression zone is injected with NaCl solution, and the lower solution tank corresponding to the tension zone is injected with NaOH solution. This setting is to simulate the migration of chloride ions from the compression zone to the tension zone.

[0060] 2. Solution injection operation: a. Use a graduated cylinder or pipette to accurately measure the required concentration of NaCl solution and NaOH solution. b. Slowly inject the solution into the corresponding solution tank, ensuring uniform distribution and no air bubbles. c. Check the sealing of the solution tank to ensure that the solution does not leak during the experiment.

[0061] Step S20C, determine the chloride ion diffusion direction according to the solution configuration, as follows: under tensile load, chloride ions migrate from the tension zone containing NaCl solution to the compression zone containing NaOH solution, with the diffusion direction from the tension side to the compression side; under compression load, chloride ions migrate from the compression zone containing NaCl solution to the tension zone containing NaOH solution, with the diffusion direction from the compression side to the tension side.

[0062] The overall process is as follows:

[0063] 1. Determine the diffusion direction based on the load type: a. Tensile load: Chloride ions migrate from the tensile zone containing NaCl solution to the compression zone containing NaOH solution, the diffusion direction is from the tensile side to the compression side. b. Compression load: Chloride ions migrate from the compression zone containing NaCl solution to the tensile zone containing NaOH solution, the diffusion direction is from the compression side to the tensile side. 2. Record the diffusion direction: Record the determined chloride ion diffusion direction in the experiment log for subsequent analysis and verification.

[0064] Example: Assume the test block is under tensile load, according to the solution configuration of step S20B, the tensile zone (lower side) is injected with NaCl solution, and the compression zone (upper side) is injected with NaOH solution. Therefore, chloride ions will migrate from the tensile zone (lower side) to the compression zone (upper side), and the diffusion direction is from the tensile side to the compression side. This direction is recorded in the experiment log.

[0065] Step S20D, store the diffusion direction in association with the test block stress type and solution configuration parameters.

[0066] Wherein, the association storage: systematically record and store the key parameters in the experiment (such as load type, solution configuration, diffusion direction, etc.) for subsequent analysis and verification. Parameter record: detailed record of various parameters in the experiment, including load type, solution type, concentration, diffusion direction, etc.

[0067] Process the test block and spray the preset color developing agent, measure the chloride ion penetration depth, including:

[0068] Step S301, rinse the test block clean and dry, cut the test block along the chloride ion diffusion direction to get the cross section.

[0069] Wherein, the chloride ion diffusion direction: the migration path of chloride ions in the concrete test block, determined by the experiment setup. Cross section: the plane perpendicular to the diffusion direction obtained by cutting the test block along the chloride ion diffusion direction, used to observe the chloride ion penetration depth.

[0070] The overall process is as follows: Test block cleaning: After the experiment, remove the test block from the loading device, rinse the surface of the test block with clean water to remove surface residues and impurities. Use a clean cloth or paper towel to dry the surface of the test block to ensure that there is no water residue on the surface.

[0071] Step S302, spray the cross section with AgNO3 color developing agent of a predetermined concentration, and let it stand for a predetermined time to allow the chloride ions to react with silver ions to form a white silver chloride precipitate boundary.

[0072] AgNO3 developer: Silver nitrate solution used to react with chloride ions to form insoluble silver chloride (AgCl) precipitate, indicating the diffusion boundary of chloride ions. Preset concentration: The concentration of AgNO3 solution determined according to experimental design, usually 0.1 mol / L. Resting time: After spraying the developer, the test block needs to be rested for a certain period of time to ensure that the chloride ions fully react with the silver ions to form a clear precipitate boundary.

[0073] The overall process is as follows: 1. Spray the developer: Use a sprayer or brush to evenly spray the AgNO3 solution of the preset concentration (such as 0.1 mol / L) on the cross-section of the test block. Ensure that the entire cross-section is covered without missing areas. 2. Rest the reaction: Place the sprayed test block in a well-ventilated environment and rest for a preset time (such as 30 minutes) to ensure that the chloride ions fully react with the silver ions. During the resting process, avoid external interference such as vibration or airflow.

[0074] Step S303, measure the vertical distance from the surface of the test block to the precipitate boundary at multiple points along the cross-section, and take the average value as the chloride ion penetration depth.

[0075] Penetration depth: The maximum distance that chloride ions diffuse in the concrete test block, usually determined by measuring the AgCl precipitate boundary formed by the reaction of chloride ions with AgNO3. Multi-point measurement: Measure at multiple points on the cross-section of the test block to ensure the accuracy and reliability of the data. Average value: Take the average of the multi-point measurement data to reduce measurement error and obtain a more accurate penetration depth.

[0076] The overall process is as follows: 1. Select measurement points: Select multiple measurement points on the cross-section of the test block, usually evenly distributed points to ensure comprehensive measurement. For example, select measurement points at the center, corners and middle positions of the cross-section. 2. Measure the penetration depth: Use a vernier caliper or digital caliper to measure the vertical distance from the surface of the test block to the AgCl precipitate boundary. Record the penetration depth data of each measurement point. 3. Calculate the average value: Add the penetration depth data of all measurement points and then divide by the number of measurement points to obtain the average penetration depth.

[0077] Step S304, stratified sampling according to the penetration depth range, detecting the chloride ion concentration of each layer, establishing the concentration gradient distribution.

[0078] Stratified sampling: According to the chloride ion penetration depth range, divide the cross-section of the test block into multiple layers and sample from each layer. Chloride ion concentration detection: Determine the concentration of chloride ions in the sample by chemical analysis method, usually using ion chromatography or potentiometric titration. Concentration gradient distribution: Describe the concentration change of chloride ions at different depths in the test block, usually represented by a curve of concentration versus depth.

[0079] The overall process is as follows: 1. Determine the delamination range: According to the depth range of chloride ion penetration, divide the cross section of the test block into multiple layers. For example, if the penetration depth is 10 mm, the test block can be divided into 0-2 mm, 2-4 mm, 4-6 mm, 6-8 mm, 8-10 mm, etc. 2. Layered sampling: Use drilling tools or cutting tools to take appropriate amount of concrete samples from each layer. Ensure that the integrity of the sample is not damaged during sampling. Label and save each layer sample separately for subsequent analysis. 3. Chloride ion concentration detection: Use ion chromatography or potentiometric titration method to determine the chloride ion concentration in each layer sample. Record the chloride ion concentration data of each layer. 4. Establish concentration gradient distribution: According to the chloride ion concentration data of each layer, draw the curve of concentration change with depth, and establish the concentration gradient distribution.

[0080] After calculating the chloride ion diffusion coefficient under different loading degrees and different stress types according to the combined experimental parameters, it also includes the steps of obtaining deterioration parameters and simulating structure performance, as follows:

[0081] Step SA00, collect the crack distribution image of the test block cross section and extract the crack geometric features, derive the acoustic emission signals recorded during the experiment and analyze the acoustic emission characteristic parameters, conduct uniaxial stress-strain curve experiment on the test block and obtain the mechanical performance parameters.

[0082] Among them, the crack distribution image: the visual representation of the position, length, width, etc. of the cracks on the test block cross section. Crack geometric features: geometric parameters such as length, width, distribution density of cracks. Acoustic emission signal: acoustic wave signal generated due to the generation and expansion of microcracks during the experiment.

[0083] Acoustic emission characteristic parameters: energy, amplitude, duration, etc. of acoustic emission signal. Uniaxial stress-strain curve: a curve describing the stress and strain relationship of the material under uniaxial stress condition. Mechanical performance parameters: material performance parameters obtained from the uniaxial stress-strain curve, such as elastic modulus, yield strength, ultimate strength, etc.

[0084] The complete process is as follows: 1. Collecting crack distribution images: Use a high-resolution industrial camera (≥20 million pixels) to take pictures of the crack distribution images of the test block cross section. Extract the geometric features of the cracks, including the length, width and distribution density of the cracks, through image processing software (such as MATLAB or OpenCV). 2. Derive acoustic emission signals and analyze: Derive acoustic emission signals from experimental records, with a signal sampling frequency of 1 MHz. Use acoustic emission analysis software (such as AEwin) to analyze acoustic emission signals and extract acoustic emission characteristic parameters, including energy count, amplitude and duration. 3. Uniaxial stress-strain curve experiment: Perform uniaxial compression or tension experiment on the test block and record the stress-strain curve. Obtain mechanical property parameters such as elastic modulus, yield strength, ultimate strength, etc. from the stress-strain curve.

[0085] Step SB00, based on the mechanical property parameters, substitute the concrete damage plasticity related formula to calculate the tensile and compressive damage evolution parameters of the deteriorated part.

[0086] Mechanical property parameters: Material performance parameters obtained from the uniaxial stress-strain curve, such as elastic modulus, yield strength, ultimate strength, etc. Damage evolution parameters: Parameters describing the accumulation and development of material damage during loading, commonly used in concrete damage plasticity constitutive models. Concrete damage plasticity constitutive model (CDP): A mathematical model describing the damage and plastic deformation of concrete during loading, usually including damage variables and plastic flow rules.

[0087] The complete process is as follows: 1. Obtain mechanical property parameters: Obtain elastic modulus (E), yield strength ( ), ultimate strength ( ), etc. from the uniaxial stress-strain curve experiment.

[0088] 2. Calculate damage evolution parameters: Use the relevant formulas of the concrete damage plasticity constitutive model (CDP) to substitute the mechanical property parameters and calculate the tensile and compressive damage evolution parameters. For example, the tensile damage evolution parameter ( ) and the compressive damage evolution parameter ( ) can be calculated by the following formulas:

[0089] ;

[0090] ;

[0091] Where, is the current stress, is the ultimate strength, is the yield strength, and n is the hardening index.

[0092] Record the damage evolution parameters: record the calculated damage evolution parameters in the experiment log for subsequent analysis.

[0093] Step SC00, construct a multi-dimensional feature matrix containing crack characteristics, acoustic emission parameters, residual strength and chloride ion diffusion coefficient, establish an end-to-end mapping relationship between the feature matrix and the CDP parameter through a deep learning network, and obtain the damage factor by inversion.

[0094] Among them, the multi-dimensional feature matrix: a matrix containing multiple feature parameters, used to describe the degradation state of the test block. The feature parameters include crack geometric characteristics, acoustic emission feature parameters, residual strength and chloride ion diffusion coefficient, etc. Deep learning network: a machine learning model based on neural network, used to learn complex mapping relationship from a large amount of data. CDP parameter: parameter in concrete damage plasticity constitutive model (CDP), used to describe the damage and plastic behavior of concrete.

[0095] The overall process can refer to steps SC10 to SC50, which will not be repeated here.

[0096] Step SD00, fuse the damage factor obtained by inversion with the calculated damage evolution parameters, combine the concentration gradient and penetration depth distribution corresponding to the chloride ion diffusion coefficient, and form the comprehensive material parameters of the degraded part.

[0097] Among them, the damage factor: a parameter describing the degradation degree of the test block obtained by deep learning network inversion. Damage evolution parameter: a parameter describing the accumulation and development of damage calculated based on mechanical property parameters. Comprehensive material parameter: a group of parameters describing the material performance of the degraded part formed after fusing the damage factor and the damage evolution parameter, used for subsequent structure performance simulation.

[0098] The overall process can refer to steps SD10 to SD60, which will not be repeated here.

[0099] Step SE00, establish a structure service mechanics model, assign the fused comprehensive material parameters to the structure service mechanics model according to the actual tension zone and compression zone of the structure, and set the boundary conditions and sustained load working conditions consistent with the preset structure service scenario.

[0100] Among them, the structure service mechanics model: a mathematical model used to simulate the mechanical behavior of the structure under actual service conditions, usually based on finite element analysis. Comprehensive material parameter: a parameter that fuses the damage factor, damage evolution parameter, chloride ion diffusion coefficient and other information, used to describe the material performance of the degraded part. Finite element analysis: a numerical analysis method used to solve complex mechanical problems, which approximates the solution by dividing the structure into a finite number of elements.

[0101] The complete process is as follows: 1. Establish the structural service mechanics model: use finite element analysis software (such as Abaqus, ANSYS) to establish the geometric model of the structure. According to the actual stress condition of the structure, set the boundary conditions and load conditions. 2. Assign comprehensive material parameters: assign the comprehensive material parameters obtained in step SD00 to the corresponding deterioration parts in the structural model. Ensure that the material parameters of different parts in the model are consistent with the actual service conditions. 3. Verify the convergence of the model: check whether the meshing of the finite element model is reasonable, and ensure the convergence of the model. Adjust the grid density or element type to improve the accuracy and convergence of the model.

[0102] Example: Suppose we need to simulate the mechanical behavior of a tunnel structure under service conditions. First, use Abaqus software to establish the geometric model of the tunnel structure, including the shape, size and support structure of the tunnel. According to the actual stress condition of the tunnel, set the boundary conditions (such as ground stress, groundwater pressure) and load conditions (such as vehicle load). Then, assign the comprehensive material parameters obtained in step SD00 to the corresponding deterioration parts in the tunnel structure. Finally, check whether the meshing of the finite element model is reasonable, and adjust the grid density to ensure the convergence of the model.

[0103] Step SF00, after verifying the convergence of the structural service mechanics model, submit the model for mechanical performance calculation; based on the calculation results, analyze the bearing capacity degradation law, stiffness attenuation characteristics and stress distribution state of the structure, compare the calculation results with the experimental measured data of the test block under the same conditions to verify the accuracy of the model, if the error ≤ preset threshold, complete the simulation of the mechanical performance degradation of the concrete structure.

[0104] Among them, model convergence: refers to whether the finite element model can stably reach the solution state in the calculation process, usually verified by checking the meshing and stability of the calculation results. Bearing capacity degradation: the decrease of bearing capacity of the structure due to material degradation during service. Stiffness attenuation: the decrease of stiffness of the structure due to material degradation during service. Stress distribution: the stress distribution of the structure under stress conditions, usually obtained by finite element analysis.

[0105] The complete process is as follows: 1. Verify model convergence: Check if the meshing of the finite element model is reasonable, and ensure the convergence of the model. Adjust the grid density or element type to improve the accuracy and convergence of the model. Run the model calculation and check if the results are stable to ensure the convergence of the model. 2. Calculate mechanical performance indicators: Run the finite element model to calculate the bearing capacity, stiffness and stress distribution of the structure. Analyze the changes in bearing capacity degradation, stiffness attenuation and stress distribution to evaluate the degree of deterioration of the structure. 3. Result analysis and verification: Compare the calculation results with experimental data or theoretical values to verify the accuracy of the model. If there is a large deviation between the calculation results and experimental data, adjust the model parameters or re-mesh the grid and re-calculate. The preset threshold can be 5%.

[0106] Assuming a finite element model of a tunnel structure has been established and comprehensive material parameters have been assigned. First, check if the meshing of the model is reasonable, ensuring that the grid density is high enough to capture the details of the structure. Run the model calculation to obtain the bearing capacity, stiffness and stress distribution of the tunnel structure. Analyze the results and find that the bearing capacity of the tunnel structure has decreased by 20%, the stiffness has decreased by 30%, and the stress distribution has shown a clear concentration area. Compare these calculation results with experimental data and find that they are basically consistent, verifying the accuracy of the model.

[0107] Through the deep learning network, an end-to-end mapping relationship between the feature matrix and the CDP parameters is established, and the damage factors are obtained by inversion, including:

[0108] Step SC10, principal component analysis is used to reduce the dimensionality of the multi-dimensional feature matrix to remove redundant information.

[0109] Wherein, the multi-dimensional feature matrix: a matrix containing multiple feature parameters, used to describe the deterioration state of the test block, the feature parameters include crack geometric characteristics, acoustic emission feature parameters, residual strength and chloride ion diffusion coefficient, etc. Principal component analysis (PCA): a statistical method that projects data into a new coordinate system through linear transformation, maximizing the variance of data in the new coordinate system, thereby removing redundant information and reducing data dimensionality. Redundant information: parts of data that are repeated or do not provide additional information, which can usually be removed by dimensionality reduction methods.

[0110] The complete process is as follows: 1. Construct a multi-dimensional feature matrix: collect crack geometric features (such as crack length, width, distribution density), acoustic emission feature parameters (such as energy count, amplitude, duration), residual strength and chloride ion diffusion coefficient, etc. Organize these data into a multi-dimensional feature matrix, each row represents a sample, and each column represents a feature parameter. 2. Apply principal component analysis (PCA): use PCA algorithm to reduce the dimension of the multi-dimensional feature matrix. Select the number of principal components to be retained, usually the number of principal components to be retained should be able to explain most of the variance of the data (such as more than 95%). The reduced dimension data is used as the input of the subsequent deep learning network.

[0111] Step SC20, construct an end-to-end deep learning network, the input layer receives the reduced multi-dimensional feature matrix, the hidden layer adopts a transfer learning strategy, pre-train the network initial weight based on the mechanical-erosion dataset of standard concrete specimens, and adapt the feature dimension by dynamically adjusting the number of hidden layer neurons, the output layer is directly mapped to the damage factor.

[0112] Wherein, end-to-end deep learning network: a deep learning model that directly maps from input data to output results without manually designing feature extraction process. Transfer learning strategy: use pre-trained model weights as initial weights to speed up the training process and improve model performance. Mechanical-erosion dataset of standard concrete specimens: contains experimental data of standard concrete specimens under different mechanical and erosion conditions, used for pre-training model. Number of hidden layer neurons: the number of neurons in the hidden layer of the deep learning network, which affects the complexity and fitting ability of the model.

[0113] The complete process is as follows: 1. Construct an end-to-end deep learning network: design a deep learning network, the input layer receives the reduced multi-dimensional feature matrix, and the output layer is directly mapped to the damage factor. The hidden layer adopts a transfer learning strategy, pre-trains the network initial weight based on the mechanical-erosion dataset of standard concrete specimens. Dynamically adjust the number of hidden layer neurons to adapt to the feature dimension and improve the performance of the model. 2. Pre-train the network: use the mechanical-erosion dataset of standard concrete specimens to pre-train the network. During the pre-training process, adjust the number of hidden layer neurons to ensure that the model can effectively learn the features in the data. 3. Optimize network structure: according to the pre-training result, further optimize the network structure, such as adjusting the number of layers, neurons, etc. Ensure that the network structure can adapt to the dimension of the reduced feature matrix.

[0114] Example: Suppose the dimension of the reduced feature matrix is 3, and a simple end-to-end deep learning network is designed, including an input layer, two hidden layers, and an output layer. The input layer receives 3 features, the first hidden layer has 10 neurons, the second hidden layer has 5 neurons, and the output layer outputs the damage factor. Pre-training is performed using a mechanical-erosion dataset of standard concrete specimens. Suppose the dataset contains 1000 samples, each with 3 features and a corresponding damage factor. During pre-training, the number of hidden layer neurons is dynamically adjusted, and finally the first hidden layer has 12 neurons and the second hidden layer has 6 neurons.

[0115] Step SC30, introduce a preset two-stage optimization mechanism for the constructed deep learning network. For details, refer to steps SC31 to SC32, which are not repeated here.

[0116] Step SC40, during the training of the deep learning network, convert the uniaxial tension / compression constitutive equation of concrete into network constraints, and add a physical constraint penalty term to the loss function. When the predicted damage factor deviates from the mechanical law, increase the penalty weight according to the deviation degree.

[0117] where the physical constraint penalty term is an additional term added to the loss function to ensure that the model's output conforms to the physical law, such as the mechanical performance of concrete. Loss function: a function used to evaluate the difference between the model's predicted value and the true value, usually minimized during training. Deviation degree: the difference between the model's predicted value and the physical law, used to adjust the weight of the penalty term.

[0118] The overall process is as follows: 1. Define physical constraints: define physical constraints based on the mechanical performance of concrete. For example, the damage factor of concrete should be between 0 and 1, and should increase with stress. Convert these physical constraints into mathematical expressions, such as: 0≤D≤1, where D is the damage factor and σ is the stress. 2. Construct the loss function: add a physical constraint penalty term to the loss function to ensure that the model's output conforms to the physical law. The loss function can be represented as: where is the data-driven loss term, is the physical constraint penalty term, and λ is the weight of the penalty term. Dynamically adjusting the penalty weight includes: dynamically adjusting the weight of the penalty term according to the deviation degree of the model's predicted value from the physical law. If the predicted value deviates from the physical law more, increase the penalty weight; if the deviation is less, reduce the penalty weight. 3. Train the model: use an optimization algorithm (such as the Adam optimizer) to minimize the loss function and train the deep learning network. During training, monitor whether the model's output conforms to the physical law to ensure that the model's predicted value is reasonable.

[0119] Example: Suppose a end-to-end deep learning network has been built, the input layer receives 3 features, two hidden layers have 12 and 6 neurons respectively, and the output layer outputs damage factor. During training, define the physical constraint condition as the damage factor should be between 0 and 1, and increase with stress. Assume the initial weight λ = 0.1, during training, if the model predicted damage factor exceeds the range of 0 to 1, increase the value of λ; if the predicted value conforms to the physical law, keep λ unchanged.

[0120] Step SC50, using the deep learning network model trained by two-stage optimization and physical constraint, infer the input reduced dimension feature matrix, output the damage factor that meets the accuracy requirement and conforms to the physical law. And associate the confidence score of the damage factor corresponding to the feature matrix.

[0121] Among them, the confidence score: the confidence degree of the model to the prediction result, usually expressed through probability or uncertainty measure.

[0122] The complete process is as follows: 1, prepare input data: the feature matrix after dimension reduction is taken as input data, ensure that the format of input data is consistent with that during training. 2, model inference: use the deep learning network model trained by two-stage optimization (NSGA-Ⅱ algorithm and Bayesian regularization algorithm) to infer the input data. The model outputs the predicted damage factor. 3, calculate the confidence score: through the uncertainty estimation or confidence interval calculation of the model, assign a confidence score to each predicted damage factor. Confidence score can be based on the output probability distribution of the model or estimated by Monte Carlo method. 4, result output: output the predicted damage factor and its corresponding confidence score. Store or display the results for subsequent analysis and verification.

[0123] Two-stage optimization mechanism includes:

[0124] Step SC31, the first stage adopts NSGA-Ⅱ algorithm for multi-objective optimization, taking the error sum of squares of CDP parameter prediction value and experimental measured value as accuracy index, and network iteration number as efficiency index, outputting accuracy-efficiency Pareto optimal solution.

[0125] Among them, NSGA-II algorithm: a non-dominated sorting based multi-objective genetic algorithm, used to optimize multiple objective functions simultaneously, outputting Pareto optimal solution. Pareto optimal solution: in multi-objective optimization, the solution that cannot improve a certain objective without deteriorating other objectives. Accuracy index: error sum of squares of CDP parameter prediction value and experimental measured value, used to evaluate the prediction accuracy of the model. Efficiency index: network iteration number, used to evaluate the training efficiency of the model.

[0126] The complete process is as follows: 1. Define optimization objectives: precision indicators: the sum of squared errors between the predicted values of CDP parameters and the actual measured values, i.e. ; wherein, is the model-predicted i-th damage factor, is the experimentally measured i-th damage factor, and n is the sample size.

[0127] Efficiency indicators: the number of network iterations, i.e., the number of iterations required for the model to converge. 2. Initialize the population: randomly generate a set of initial population, each individual represents a set of network parameters (such as weights and biases). 3. Non-dominated sorting: non-dominated sorting of individuals in the population, and individuals are divided into different non-dominated layers. In the same non-dominated layer, individuals are sorted according to the crowding distance to maintain the diversity of the population. 4. Genetic operation: selection operation: select individuals from the population for crossover and mutation operations. Crossover operation: combine the parameters of two parent individuals to generate new offspring individuals. Mutation operation: randomly perturb the parameters of individuals to introduce new genetic variations. 5. Iterative optimization: repeat non-dominated sorting and genetic operation until the preset number of iterations or population convergence is reached. Output the Pareto optimal solution, i.e., the solution set that achieves the best balance between precision and efficiency.

[0128] Example: Suppose we have a dataset containing 100 samples, each containing 3 features and corresponding damage factors. We use the NSGA-II algorithm for multi-objective optimization, aiming to find model parameters with high precision and few iterations. The initial population size is 50, the crossover probability is 0.9, and the mutation probability is 0.1. After 100 generations of iteration, we obtain a set of Pareto optimal solutions, with the optimal solution having an error sum of 0.05 and an iteration number of 80.

[0129] Step SC32, based on the optimal solution, enter the second stage, dynamically adjust the network weights through the Bayesian regularization algorithm to reduce the risk of overfitting.

[0130] wherein, Bayesian regularization algorithm: a regularization method based on Bayesian theory, used to dynamically adjust network weights to reduce the risk of overfitting. Overfitting: the phenomenon that the model performs well on the training set but poorly on the test set or actual data. Weight adjustment: adjust the weights and biases of the network to optimize the performance of the model.

[0131] The overall process is as follows: 1. Initialize network weights: use the optimal solution obtained in the first stage (NSGA-II algorithm) to initialize network weights. 2. Bayesian regularization: during training, introduce a Bayesian regularization term to dynamically adjust network weights. The regularization term usually includes the prior distribution of weights and the likelihood function, which is used to control the complexity of weights. 3. Dynamically adjust weights: in each iteration, update network weights according to the Bayesian regularization term. By adjusting the regularization parameter, balance the fitting ability and generalization ability of the model. 4. Monitor the training process: use the validation set to monitor the performance of the model to ensure that the model does not overfit. If the loss function of the validation set starts to rise, training can be stopped in advance. 5. Output optimized weights: after training, output the network weights optimized by Bayesian regularization.

[0132] Example: Suppose in the first stage (NSGA-II algorithm), a set of optimal network weights has been obtained. In the second stage, these weights are further optimized using the Bayesian regularization algorithm. Suppose the regularization parameter is λ = 0.01, and the formula for updating weights in each iteration during training is: ; where θ is the network weight, α is the learning rate, and L is the loss function. By dynamically adjusting the weights, ensure that the model performs well on both the training set and the validation set.

[0133] The damage factor obtained by inversion is combined with the calculated damage evolution parameters, combined with the concentration gradient and penetration depth distribution corresponding to the chloride ion diffusion coefficient, to form the comprehensive material parameters of the deterioration site, including:

[0134] Step SD10, collect the damage factor obtained by inversion, the tensile / compressive damage evolution parameters calculated and obtained, and the concentration gradient and penetration depth distribution data corresponding to the chloride ion diffusion coefficient, based on the inversion error, theoretical fitness and measurement accuracy of each parameter, determine the fusion weight of each parameter.

[0135] Where fusion weight: when fusing different parameters, according to the inversion error, theoretical fitness and measurement accuracy of each parameter, assign a weight to each parameter. Inversion error: the difference between the model prediction value and the experimental measured value. Theoretical fitness: the matching degree between the model prediction value and the theoretical value.

[0136] Measurement accuracy: the accuracy and reliability of experimental measurement.

[0137] The complete process is as follows: 1. Collect data: collect the damage factor obtained by inversion, the tensile / compressive damage evolution parameters calculated and obtained, and the concentration gradient and penetration depth distribution data corresponding to the chloride ion diffusion coefficient. 2. Evaluate the reliability of each parameter: calculate the inversion error of each parameter, for example: ; evaluate the theoretical fitness of each parameter, e.g., whether the damage factor conforms to physical laws. Evaluate the measurement accuracy of each parameter, e.g., the measurement error range of the concentration gradient. 3. Determine the fusion weight: assign a fusion weight to each parameter according to the inversion error, theoretical fitness, and measurement accuracy. The assignment of weights can be based on the following rules: parameters with small inversion errors, high theoretical fitness, and high measurement accuracy are assigned higher weights. Parameters with large inversion errors, low theoretical fitness, and low measurement accuracy are assigned lower weights.

[0138] For example, if the inversion error of the damage factor is 5%, the theoretical fitness is 90%, and the measurement accuracy is 95%, a higher weight can be assigned; if the inversion error of the concentration gradient is 10%, the theoretical fitness is 80%, and the measurement accuracy is 90%, a lower weight can be assigned.

[0139] Step SD20, parameter adaptation for structure tension zone and compression zone division, the tension zone is related to the damage factor through the quantitative relationship between crack density and penetration depth, and the compression zone is matched with the damage evolution parameter through the corresponding relationship between stiffness degradation rate and concentration gradient. Among them, the tension zone: the area of the structure under tensile stress under the action of load. Compression zone: the area of the structure under compression stress under the action of load. Crack density: the number of cracks per unit area, usually used to evaluate the damage degree of the tension zone. Stiffness degradation rate: the degree of reduction of structural stiffness, usually used to evaluate the damage degree of the compression zone. Quantitative relationship: the relationship between crack density and damage factor, stiffness degradation rate and damage evolution parameter is quantified for parameter adaptation.

[0140] The complete process is as follows: 1. Determine the tension zone and compression zone: according to the stress analysis of the structure, determine the specific location of the tension zone and the compression zone. 2. Quantify the relationship between crack density and damage factor: in the tension zone, relate the damage factor through the quantitative relationship between crack density and damage factor. For example, the following formula can be used: ; wherein, is the damage factor of the tension zone, f is the quantification function, which can be linear or nonlinear function. 3. Quantify the relationship between stiffness degradation rate and damage evolution parameter: in the compression zone, match the damage evolution parameter through the corresponding relationship between stiffness degradation rate and damage evolution parameter. For example, the following formula can be used: ;

[0141] wherein, is the damage evolution parameter of the compression zone, g is the quantification function, which can be linear or nonlinear function. 3. Parameter adaptation: according to the quantitative relationship, adapt crack density and damage factor, stiffness degradation rate and damage evolution parameter to ensure the rationality and consistency of parameters in the tension zone and the compression zone.

[0142] Step SD30, based on the partition adaptation relationship, the mechanical parameters and the corresponding chloride ion parameters are weighted according to the preset weight, the erosion influence coefficient is introduced to realize quantitative coupling, and the partition fusion parameters are obtained, wherein the mechanical parameters include the damage factor of the tension zone and the damage evolution parameter of the compression zone, and the chloride ion parameters include the penetration depth and the concentration gradient.

[0143] Wherein, the partition adaptation relationship: according to the different characteristics of the tension zone and the compression zone, the corresponding mechanical parameters and chloride ion parameters are adapted. Mechanical parameters: including the damage factor of the tension zone and the damage evolution parameter of the compression zone. Chloride ion parameters: including penetration depth and concentration gradient. Erosion influence coefficient: used to quantify the influence degree of chloride ion erosion on mechanical properties. Weighted average: according to the preset weight, the different parameters are weighted and summed to obtain the comprehensive material parameters.

[0144] The complete process is as follows: 1. Determine the preset weight: according to the fusion weight determined in step SD10, the weights of the mechanical parameters and the chloride ion parameters are allocated. For example, the weight of the damage factor of the tension zone is 0.3, the weight of the damage evolution parameter of the compression zone is 0.28, the weight of the penetration depth is 0.05, and the weight of the concentration gradient is 0.12. 2. Calculate the partition fusion parameters: in the tension zone, the damage factor and the penetration depth are weighted according to the weight to obtain the tension zone fusion parameters, and the specific formula is as follows: .

[0145] In the compression zone, the damage evolution parameter and the concentration gradient are weighted according to the weight to obtain the compression zone fusion parameters, and the specific formula is as follows: .

[0146] Wherein, and are the fusion parameters of the tension zone and the compression zone respectively, , is the weight of the mechanical parameter, , is the weight of the chloride ion parameter, is the damage evolution parameter of the compression zone, is the damage factor of the tension zone, d is the penetration depth, and C is the concentration gradient.

[0147] 3. Introduce the erosion influence coefficient: introduce the erosion influence coefficient γ to quantify the influence of chloride ion erosion on mechanical properties. Correct the fusion parameters: ; ; wherein, and are the erosion influence coefficients of the tension zone and the compression zone respectively.

[0148] 4. Output the partition fusion parameters: output the corrected tension zone and compression zone fusion parameters for subsequent spatial mapping and comprehensive material parameter construction.

[0149] Step SD40, spatially map the zoned fusion parameters by penetration depth, form a gradientized parameter that varies continuously along the erosion depth by interpolation, and fit the actual deterioration spatial distribution.

[0150] Wherein, zoned fusion parameters: the comprehensive parameters of the tensile zone and the compression zone after being weighted and modified by the erosion influence coefficient. Spatial mapping: distribute the parameter values according to their spatial positions in the structure to form a continuous parameter field. Interpolation: estimate the value of unknown data points between known data points by mathematical methods to form a continuous parameter distribution.

[0151] The overall process is as follows:

[0152] 1. Determine the spatial position: determine the spatial position of the tensile zone and the compression zone according to the geometric shape of the structure and the force analysis. Associate the fusion parameters of each region with the corresponding spatial coordinates.

[0153] 2. Spatial mapping: spatially map the zoned fusion parameters by penetration depth to form a parameter distribution along the erosion depth. Use interpolation methods (such as linear interpolation, spline interpolation, etc.) to estimate the value of unknown data points between known data points to form a continuous parameter distribution.

[0154] 3. Generate gradientized parameters: generate gradientized parameters that vary continuously along the erosion depth by interpolation methods to ensure that the parameter distribution fits the actual deterioration spatial distribution.

[0155] Step SD50, verify the gradientized parameters based on the uniaxial tension / compression constitutive equation of concrete, and correct the parameters that deviate from the physical law according to the constraints.

[0156] Wherein, gradientized parameters: parameters that vary continuously along the erosion depth, reflecting the spatial distribution of material properties. Uniaxial tension / compression constitutive equation: a mathematical model that describes the stress-strain relationship of concrete under uniaxial tension or compression. Physical law: the basic physical principles that material properties should comply with, such as the damage factor should be between 0 and 1, and increase with stress. Constraint correction: adjust the parameters that deviate from the physical law to make them comply with the physical law.

[0157] 1. Define the uniaxial tension / compression constitutive equation: use the uniaxial tension / compression constitutive equation of concrete to verify the gradientized parameters. For example, the tensile constitutive equation can be expressed as:

[0158] ;

[0159] The compression constitutive equation can be expressed as: ; wherein, : the tensile stress per unit area of concrete under uniaxial tension, : Stiffness index of concrete in tension elastic stage (stress-strain proportional stage), i.e. the ratio of stress to strain in elastic stage, : Deformation of concrete in tension (dimensionless or expressed in micro-strain με), which is the degree of elongation deformation under tensile stress, : Dimensionless parameter describing the degradation of concrete in tension. : Compressive stress per unit area that concrete bears under uniaxial compression state; : Stiffness index of concrete in compression elastic stage (stress-strain proportional stage), : Deformation of concrete in compression (dimensionless), which includes elastic strain (without obvious plastic deformation) and plastic / damage strain, : Dimensionless parameter quantifying the weakening degree of mechanical properties due to "micro-crack propagation, aggregate-mortar interface slip, internal defect accumulation" and other damages during the compression process of concrete.

[0160] 2. Verify the gradientized parameters: Substitute the gradientized parameters into the uniaxial tension / compression constitutive equation and check if they meet the physical laws. For example, check if the damage factor is between 0 and 1 and increases with stress.

[0161] 3. Constraint correction: If the gradientized parameters deviate from the physical laws, correct them according to the constraints of the constitutive equation. For example, if the damage factor exceeds 1, correct it to 1; if the damage factor is less than 0, correct it to 0.

[0162] 4. Output the corrected gradientized parameters: Output the gradientized parameters after constraint correction to ensure they meet the physical laws.

[0163] Step SD60, associate the corrected gradientized parameters with the tension / compression partition identification and penetration depth coordinates to form comprehensive material parameters that include mechanical degradation, chloride ion erosion and spatial distribution characteristics.

[0164] Among them, comprehensive material parameters: parameters that integrate mechanical performance degradation, chloride ion erosion and spatial distribution characteristics, used to comprehensively describe the performance changes of materials under service conditions. Mechanical degradation: the degradation of material performance during loading, such as strength reduction, stiffness degradation, etc. Chloride ion erosion: the chemical erosion of chloride ions on materials, leading to performance decline. Spatial distribution characteristics: the spatial variation of material performance, usually related to erosion depth.

[0165] The overall process is as follows: 1. Integrating the corrected gradient parameters: associate the corrected gradient parameters with the tensile / compressive partition identification and the penetration depth coordinates. For example, for each depth point in the tensile and compressive zones, record the corresponding gradient parameter value. 2. Constructing comprehensive material parameters: integrate the mechanical degradation parameters (such as damage factor, stiffness degradation rate), chloride ion erosion parameters (such as penetration depth, concentration gradient) and spatial distribution characteristics (such as depth coordinates) into a comprehensive parameter system. 3. Output the integrated comprehensive material parameters as tables or data files for subsequent structural performance analysis and simulation.

[0166] In addition, after completing the mechanical performance degradation simulation, a concrete degradation simulation method under sustained load-chloride ion coupling also includes a simulation-prediction-verification closed-loop analysis step, which is as follows:

[0167] Step 1, build a digital twin of the structure, embed the CDP constitutive parameters (i.e. comprehensive material parameters) calibrated in the laboratory and the chloride ion diffusion model, deploy distributed optical fiber sensors, wireless chloride ion probes and three-dimensional laser scanners, real-time collect structure strain field, chloride ion concentration gradient and surface morphology data, after time-space data fusion algorithm processing, update the twin load and environmental parameters according to the preset frequency, build the initial model containing real-time degradation conditions.

[0168] Among them, the digital twin of the structure: a virtual model corresponding to the actual structure built by digital technology, used for real-time monitoring and prediction of the performance change of the structure. CDP constitutive parameters: parameters of the concrete damage plasticity constitutive model, used to describe the damage and plasticity behavior of concrete under stress.

[0169] Chloride ion diffusion model: a mathematical model describing the diffusion process of chloride ions in concrete, used to predict the distribution of chloride ion concentration. Time-space data fusion algorithm: an algorithm for fusing data at different times and spaces to improve the accuracy and completeness of the data.

[0170] The overall process is as follows: 1. Build a digital twin: embed the laboratory calibrated CDP constitutive parameters and chloride diffusion model to build a digital twin of the structure. Deploy distributed fiber optic sensors, wireless chloride probes and three-dimensional laser scanners to collect real-time data on structural strain field, chloride concentration gradient and surface topography. 2. Data processing and model updating: use spatio-temporal data fusion algorithm to process the collected data, update the twin load and environmental parameters at the preset frequency. Build an initial model containing real-time deterioration conditions to provide a basis for subsequent prediction and verification. Example: Suppose a concrete structure is exposed to a marine environment and is severely eroded by chloride ions. Real-time data is collected through the deployed sensors, processed using the spatio-temporal data fusion algorithm, and the load and environmental parameters of the twin are updated. The embedded chloride diffusion model predicts the chloride concentration distribution, and the CDP constitutive parameters describe the damage behavior of concrete. Finally, the initial model constructed can reflect the real-time deterioration state of the structure.

[0171] Step 2, build a hybrid attention mechanism LSTM prediction model, input layer fusion strain, chloride concentration, temperature and humidity and time series crack image features, use multi-head attention mechanism to dynamically allocate parameter weights, output bearing capacity degradation coefficient and crack propagation rate, train until the validation set error is less than the preset threshold, quantify the prediction uncertainty, when the predicted bearing capacity drops by more than the preset proportion and the uncertainty index is greater than the preset value, trigger an early warning.

[0172] Among them, the hybrid attention mechanism: a model that combines multiple attention mechanisms (such as self-attention and multi-head attention), used to dynamically allocate the weights of input features, improving the model's attention to important features. LSTM (Long Short-Term Memory Network): a special type of recurrent neural network (RNN) that can learn long-term dependencies, suitable for time series data prediction. Bearing capacity degradation coefficient: a parameter used to describe the reduction of structural bearing capacity over time or damage degree. Crack propagation rate: the rate of crack expansion over time or load change. Prediction uncertainty: the degree of uncertainty in the model's prediction results, usually represented by confidence intervals or probability distributions.

[0173] The complete process is as follows: 1. Construct a hybrid attention mechanism LSTM model: The input layer fuses strain, chloride ion concentration, temperature and humidity, and time series crack image features. A multi-head attention mechanism is used to dynamically allocate parameter weights, improving the model's focus on important features. The output layer outputs the bearing capacity degradation coefficient and crack propagation rate. 2. Model training and verification: Use historical data to train the hybrid attention mechanism LSTM model until the validation set error is less than the preset threshold. Quantify the prediction uncertainty to evaluate the prediction accuracy and reliability of the model. 3. Early warning mechanism: When the predicted bearing capacity decreases by more than the preset percentage and the uncertainty index is greater than the preset value, trigger the early warning. The early warning mechanism can notify relevant personnel to take timely measures to avoid further deterioration of the structure. Example: Suppose the real-time monitoring data of a concrete structure shows that the crack width increases gradually over time. Use the hybrid attention mechanism LSTM model to predict, input features include strain, chloride ion concentration, temperature and humidity, and crack image features. After model training, the predicted bearing capacity degradation coefficient is 0.8 and the crack propagation rate is 0.05 mm / month. When the predicted bearing capacity decreases by more than 20% and the uncertainty index is greater than 0.1, trigger the early warning, prompting relevant personnel to check and maintain.

[0174] Step 3, periodically obtain internal defect and composition evolution data through ultrasonic tomography and X-ray fluorescence spectrum, calculate the weighted error between measured and predicted values, if the error ≥ preset threshold, start particle swarm-Bayesian joint optimization algorithm to correct diffusion coefficient and CDP parameters (parameter space is divided according to engineering experience and set constraint boundary, after global search by particle swarm and local refinement by Bayesian, verify whether the parameters meet the physical laws of concrete deterioration, update the twin material library. Among them, internal defects: damage or defects inside the structure, such as cracks, pores, etc., affecting the integrity and durability of the structure. Composition evolution: the process of material composition changing over time or environment, such as the change of chloride ion concentration. Particle swarm-Bayesian joint optimization algorithm: an algorithm combining particle swarm optimization (PSO) and Bayesian optimization for parameter optimization. Parameter space division: divide the possible value range of parameters into multiple subintervals to facilitate the search of optimization algorithm. Physical law verification: check whether the optimized parameters meet the physical properties and behavior rules of materials.

[0175] The complete process is as follows: 1. Obtain internal defect and composition evolution data: periodically obtain internal defect and composition evolution data through ultrasonic tomography, X-ray fluorescence spectroscopy, etc. These data are used to evaluate the accuracy of model prediction and provide basis for parameter correction. 2. Calculate the weighted error: calculate the weighted error between the measured value and the predicted value to evaluate the accuracy of model prediction. 3. Start the optimization algorithm: if the weighted error is greater than the preset threshold, start the particle swarm-Bayesian joint optimization algorithm to correct the diffusion coefficient and CDP parameters. The parameter space is divided according to engineering experience and the constraint boundary is set. After global search by particle swarm and local refinement by Bayesian, it is verified whether the parameters meet the physical law of concrete deterioration.

[0176] 4. Update the twin material library: update the corrected parameters to the material library of the structure digital twin, ensuring that the twin can accurately reflect the current state of the structure.

[0177] Example: Suppose a concrete structure is found to have internal cracks through ultrasonic tomography and changes in chloride ion concentration through X-ray fluorescence spectroscopy during service. Calculate the weighted error between the measured value and the predicted value, and find that the error is greater than the preset threshold. Start the particle swarm-Bayesian joint optimization algorithm to correct the diffusion coefficient and CDP parameters. After verification, the corrected parameters meet the physical law of concrete deterioration, and are updated to the twin material library.

[0178] Step 4, if the accuracy is still not up to standard after continuous preset number of optimizations, automatically adjust the sensor configuration (such as increase the sampling frequency), trigger the unmanned aerial vehicle inspection to obtain panoramic images, use the incremental data to train the meta-learning model, until the prediction error is stable within the preset range, forming a closed-loop system.

[0179] Among them, the sensor configuration adjustment: according to the accuracy and reliability of the monitoring data, automatically adjust the sampling frequency, position and other parameters of the sensor. Unmanned aerial vehicle inspection: use unmanned aerial vehicle to collect panoramic images and obtain macro state information of the structure. Incremental data: data added during optimization, used to further train and optimize the model. Meta-learning model: a machine learning model that can quickly adapt to new tasks and new data, improving the generalization ability of the model by learning the experience of multiple tasks. Closed-loop system: a system that can automatically adjust and optimize its own parameters to ensure the stability and performance of the system.

[0180] The complete process is as follows: 1. Evaluate the optimization accuracy: if the accuracy is still not up to standard after continuous optimization, trigger the system to enter the adjustment stage. Evaluate the accuracy of the current sensor configuration and data collection. 2. Automatically adjust the sensor configuration: automatically adjust the sampling frequency or encryption scanning point density of the sensor to improve the accuracy and reliability of the data. For example, increase the sampling frequency of the sensor from once every hour to once every half hour. 3. Trigger the UAV inspection: trigger the UAV to collect panoramic images and obtain macro-state information of the structure. The image data collected by the UAV can be used to supplement the sensor data and provide more comprehensive structural state information. 4. Train the meta-learning model with incremental data: use the incremental data collected by the UAV to train the meta-learning model to further optimize the performance of the model. The meta-learning model can quickly adapt to new tasks and new data, improving the generalization ability of the model. 5. Form a closed-loop system: integrate the adjusted sensor data and image data collected by the UAV into the system to form a closed-loop system. The system automatically adjusts and optimizes its own parameters to ensure the stability and performance of the system.

[0181] The process of starting the particle swarm-Bayesian joint optimization algorithm to correct the diffusion coefficient and CDP parameters is as follows:

[0182] Step 3.1, parameter space division: according to engineering experience, divide the diffusion coefficient, damage factor and other key parameters into multiple subspaces, set reasonable upper and lower limit constraints for each subspace (such as diffusion coefficient ≤ initial value 3 times), and form parameter constraint boundaries. Among them, parameter space: the set of all possible parameter values, including diffusion coefficient, damage factor and other key parameters. Subspace: division of parameter space, divide the parameter space into multiple smaller regions to facilitate the search of the optimization algorithm. Upper and lower limit constraints: set reasonable value range for each parameter to ensure that the parameter is reasonable in physical meaning and engineering experience. The complete process is as follows: 1. Determine the key parameters: determine the key parameters to be optimized, such as diffusion coefficient, damage factor, etc. 2. Divide the parameter space: divide the parameter space into multiple subspaces according to engineering experience. Set reasonable upper and lower limit constraints for each subspace. For example, the upper limit of the diffusion coefficient can be set to 3 times the initial value. 3. Set the constraint boundary: set the upper and lower limit constraints for each subspace to form the parameter constraint boundary. For example, the upper limit of the diffusion coefficient can be set to 3 times the initial value, and the lower limit can be set to 1 / 3 of the initial value.

[0183] Step 3.2, Particle Swarm Global Search: The particle swarm global search algorithm is used to quickly locate the optimal parameter region by iteratively calculating the fitness value (measured and predicted error) and selecting the elite parameter combination. Particle Swarm Optimization (PSO) is a swarm intelligence-based optimization algorithm that simulates the foraging behavior of bird flocks to find the optimal solution. Fitness value is a function used to evaluate the quality of a solution, usually the error between the measured and predicted values. Elite parameter combination is the parameter combination that performs best during optimization, usually with the lowest fitness value. The complete process is as follows: 1. Initialize particle swarm: Randomly generate a set of particles, each representing a parameter combination, initialize position and velocity. The position of a particle corresponds to a point in the parameter space, and the velocity determines the direction and distance of the particle in space. 2. Calculate fitness value: For each particle's parameter combination, calculate the fitness value, i.e. the error between the measured and predicted values. The lower the fitness value, the closer the particle's position to the optimal solution. 3. Update particle position and velocity: Update the position and velocity of each particle according to its individual experience and group experience. Individual experience refers to the optimal position found by the particle itself, and group experience refers to the optimal position found by the entire particle swarm. 4. Select elite parameter combination: In each iteration, record the position of the particle with the lowest fitness value, i.e. the elite parameter combination. Through multiple iterations, the particle swarm gradually approaches the optimal solution, and finally selects the elite parameter combination. Example: Suppose in the particle swarm optimization algorithm, the particle swarm size is 50 and the maximum number of iterations is 100. Each particle represents a parameter combination, including the diffusion coefficient and damage factor. In each iteration, calculate the fitness value of each particle, i.e. the error between the measured and predicted values. By updating the position and velocity of the particles, the optimal solution is gradually found. Finally, the elite parameter combination with the lowest fitness value is selected.

[0184] Step 3.3, Bayesian local refinement: Based on the elite solution, a local search space is constructed, a Gaussian process regression is used to model the objective function, and the optimal parameter combination is selected by the acquisition function to achieve local fine optimization. Among them, Bayesian optimization: a global optimization algorithm based on Bayesian theory, which finds the optimal solution by constructing the prior distribution of the objective function and constantly updating the posterior distribution. Gaussian process regression: a non-parametric Bayesian method used to model and predict the value of the objective function, commonly used in Bayesian optimization. Acquisition function: a function used to determine the next sampling point, common ones include expected improvement, probability of improvement, and upper confidence bound. The complete process is as follows: 1. Constructing a local search space: based on the elite parameter combination screened by the particle swarm optimization algorithm, a local search space is constructed. The range of the local search space can be set to the neighborhood of the elite parameter combination, for example, the upper and lower limits can be set to ±10% of the elite parameters. 2. Gaussian process regression modeling: use Gaussian process regression to model the objective function, the objective function is usually the error between the measured value and the predicted value. Gaussian process regression can provide mean and variance prediction of the objective function, which is used to evaluate the uncertainty of each parameter combination. 3. Select the optimal parameter combination: use the acquisition function (such as expected improvement EI) to select the next sampling point. The acquisition function selects the parameter combination that is most likely to improve the current optimal solution according to the prediction results of the Gaussian process regression. 4. Local fine optimization: within the local search space, through multiple iterations, the optimal solution is gradually approached. In each iteration, the Gaussian process regression model is updated, and the optimal parameter combination is reselected until the convergence condition is met.

[0185] Step 3.4, Physical Constraint Verification: Verify that the optimized parameters conform to the physical laws of concrete degradation (e.g., elastic modulus degradation rate ≤ 60%). If not, adjust the search strategy and iterate again until the parameters converge stably. Physical constraint verification: Check whether the optimized parameters conform to the physical properties and behavior of the material. The elastic modulus degradation rate is the percentage by which the material's elastic modulus decreases with time or damage, often used to assess the degree of material degradation. Convergence criteria: The conditions under which the optimization algorithm stops iterating, typically based on parameter stability or error minimization. The overall process is as follows: 1. Define physical constraints: Define physical constraints based on the material's physical properties and behavior. For example, the elastic modulus degradation rate should not exceed 60%. Other possible constraints include the damage factor, which should be between 0 and 1 and increase with increasing stress. 2. Verify optimized parameters: Check whether the optimized parameters meet the physical constraints. For example, calculate the optimized elastic modulus degradation rate to ensure it does not exceed 60%. If the parameters do not meet the physical constraints, adjust the search strategy and retry the optimization. 3. Adjust the search strategy: If the parameters do not meet the physical constraints, adjust the search strategy, such as narrowing the search range or changing the optimization algorithm parameters. Re-optimize until the parameters meet the physical constraints. 4. Verify parameter stability: Verify that the optimized parameters have converged stably. If the parameters remain unstable after multiple iterations, continue to adjust the search strategy. Ensure that the parameters have converged stably after multiple iterations and meet the convergence conditions of the optimization algorithm.

[0186] Based on the same inventive concept, an embodiment of the present invention provides a simulation system for concrete degradation under continuous load-chloride ion coupling, including a memory and a processor, wherein the memory stores data that can be executed on the processor to implement the following Figure 1 Procedure of the method shown.

[0187] The embodiments of this specific implementation method are all preferred embodiments of the present application and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for simulating concrete degradation under continuous load-chloride ion coupling, characterized in that: include: A concrete test block is prepared using a preset preparation method. After standard curing and saturated solution pretreatment, the support points are adjusted and the test block is placed according to the three-point bending specification. The tensile and compressive zones of the test block are determined. The test block is loaded using a preset loading device, and the load is continuously increased until the load reaches a peak value. This peak value is recorded as the ultimate load of the test block using a preset sensor and load digital display board. A specific ratio of continuous tensile / compressive load is applied and maintained based on the ultimate load, and real-time monitoring and feedback control are performed using sensors and load digital display boards. The tensile load corresponds to the tensile zone of the test block, and the compressive load corresponds to the compressive zone of the test block. Assemble the test components based on the stress area of ​​the test block, adjust the solution according to the stress type to determine the direction of chloride ion diffusion, connect the electrodes and apply an external electric field to accelerate chloride ion migration; during this period, regularly replenish the electrode solution according to the preset solution replenishment method to maintain a stable test environment. The entire chloride ion migration stage continues until the preset experimental end requirements are met; After the experiment, the power supply was disconnected, the test pieces were processed and sprayed with the preset color developer, the chloride ion penetration depth was measured, and the chloride ion diffusion coefficients under different loading levels and different stress types were calculated based on the experimental parameters. Combining the diffusion coefficient data under different conditions, the relationship between loading degree, force type and chloride ion diffusion coefficient is established.

2. The method for simulating concrete degradation under continuous load-chloride ion coupling according to claim 1, characterized in that: Assembling test components based on the test block's stress area includes: Configure the chloride ion test components including solution tank, titanium mesh, conductive connector and fixing device, and clean and pre-treat the surface of the tensile and compressive areas of the test block; Based on the positions of the tensile and compressive areas of the pre-treated test block, the titanium mesh is laid on the stress-bearing surface and a stable conductive connection is achieved through a conductive connector. The solution tank is then positioned and assembled outside the stress-bearing area of ​​the test block using a fixing device to form a test component that is adapted to the stress-bearing area.

3. The method for simulating concrete degradation under continuous load-chloride ion coupling according to claim 1, characterized in that: Adjusting the solution to determine the direction of chloride ion diffusion based on the type of force applied includes: Determine the type of sustained load the specimen is subjected to and mark the specific spatial locations of the tensile and compressive zones of the specimen. Sustained load types include tensile and compressive loads. Based on the marked stress area, the solution is configured according to the load type, as follows: when subjected to tension load, the upper solution tank corresponding to the compression area is filled with NaOH solution, and the lower solution tank corresponding to the tension area is filled with NaCl solution; when subjected to compression load, the upper solution tank corresponding to the compression area is filled with NaCl solution, and the lower solution tank corresponding to the tension area is filled with NaOH solution; The diffusion direction of chloride ions is determined according to the solution configuration, as follows: under tensile load, chloride ions migrate from the tensile zone containing NaCl solution to the compressive zone containing NaOH solution, and the diffusion direction is from the tensile side to the compressive side; under compressive load, chloride ions migrate from the compressive zone containing NaCl solution to the tensile zone containing NaOH solution, and the diffusion direction is from the compressive side to the tensile side; The diffusion direction is associated with the test block force type and solution configuration parameters and stored.

4. The method for simulating concrete degradation under continuous load-chloride ion coupling according to claim 1, characterized in that: Treat the test block and spray the preset color developer to measure the chloride ion penetration depth including: Rinse and dry the test piece, and cut the test piece along the chloride ion diffusion direction to obtain a cross section; Spray a preset concentration of AgNO3 developer onto the cross section and let it stand for a preset time to allow chloride ions to react with silver ions to form a white silver chloride precipitation boundary; Measure the vertical distance from the specimen surface to the precipitation boundary at multiple points along the cross section, and take the average value as the chloride ion penetration depth; Sampling was carried out in layers according to the penetration depth range, the chloride ion concentration in each layer was tested, and the concentration gradient distribution was established.

5. The method for simulating concrete degradation under continuous load-chloride ion coupling according to claim 4, characterized in that: After combining experimental parameters to calculate the chloride ion diffusion coefficient under different loading levels and different stress types, the process also includes the steps of obtaining degradation parameters and simulating structural performance, as follows: Collect crack distribution images of the specimen cross section and extract crack geometric features. Export the acoustic emission signals recorded during the experiment and analyze the acoustic emission characteristic parameters. Perform uniaxial stress-strain curve experiments on the specimen and obtain mechanical property parameters. Based on the mechanical properties parameters, the damage evolution parameters of the deteriorated parts under tension and compression are calculated by substituting them into the damage plasticity constitutive formula of concrete. A multidimensional feature matrix containing crack characteristics, acoustic emission parameters, residual strength, and chloride ion diffusion coefficient was constructed. An end-to-end mapping relationship between the feature matrix and CDP parameters was established through a deep learning network, and the damage factor was obtained by inversion. The damage factor obtained by inversion is integrated with the calculated damage evolution parameters, and the concentration gradient and penetration depth distribution corresponding to the chloride ion diffusion coefficient are combined to form the comprehensive material parameters of the deteriorated part; Establish a structural service mechanics model, assign the integrated material parameters to the structural service mechanics model according to the actual tension and compression areas of the structure, and set boundary conditions and continuous load conditions consistent with the preset structural service scenario; After verifying the convergence of the structural service mechanical model, submit the model for mechanical performance calculation. Based on the calculation results, analyze the structure's bearing capacity degradation law, stiffness attenuation characteristics, and stress distribution state. Compare the calculation results with the measured data of the test block under the same conditions to verify the accuracy of the model. If the error is ≤ the preset threshold, the simulation of the mechanical performance degradation of the concrete structure is completed.

6. The method for simulating concrete degradation under continuous load-chloride ion coupling according to claim 5, characterized in that: An end-to-end mapping relationship between the feature matrix and CDP parameters is established through a deep learning network. The damage factors obtained by inversion include: Principal component analysis is used to reduce the dimensionality of the multidimensional feature matrix and remove redundant information; An end-to-end deep learning network was constructed. The input layer received a reduced multidimensional feature matrix. The hidden layer used a transfer learning strategy. The initial network weights were pre-trained based on a mechanics-erosion dataset of standard concrete specimens. The number of hidden layer neurons was dynamically adjusted to adapt to the feature dimensions. The output layer was directly mapped to damage factors. A preset two-stage optimization mechanism is introduced for optimizing the constructed deep learning network; During the deep learning network training process, the uniaxial tension / compression constitutive equation of concrete is converted into network constraints, and a physical constraint penalty term is added to the loss function. When the predicted damage factor deviates from the mechanical law, the penalty weight is increased according to the degree of deviation. Using a deep learning network model that has undergone two-stage optimization and physical constraint training, the input dimensionality reduction feature matrix is ​​inferred and a damage factor that meets the accuracy requirements and conforms to physical laws is output.

7. The method for simulating concrete degradation under continuous load-chloride ion coupling according to claim 6, characterized in that: The two-stage optimization mechanism includes: In the first stage, the NSGA-Ⅱ algorithm is used for multi-objective optimization, with the sum of squared errors between the CDP parameter prediction value and the experimental measured value as the accuracy index and the number of network iterations as the efficiency index, and the accuracy-efficiency Pareto optimal solution is output; Based on this optimal solution, we enter the second stage and dynamically adjust the network weights through the Bayesian regularization algorithm to reduce the risk of overfitting.

8. The method for simulating concrete degradation under continuous load-chloride ion coupling according to claim 5, characterized in that: The damage factor obtained by inversion is integrated with the calculated damage evolution parameters, and combined with the concentration gradient and penetration depth distribution corresponding to the chloride ion diffusion coefficient to form the comprehensive material parameters of the deteriorated part, including: Collect the damage factors obtained by inversion, the calculated tensile / compressive damage evolution parameters, and the concentration gradient and penetration depth distribution data corresponding to the chloride ion diffusion coefficient. Based on the inversion error, theoretical fitness, and measurement accuracy of each parameter, determine their respective fusion weights. Parameter adaptation is performed based on the division of the structure into tension and compression zones. The damage factor of the tension zone is associated with the quantitative relationship between crack density and penetration depth, while the damage evolution parameter of the compression zone is matched by the corresponding relationship between stiffness degradation rate and concentration gradient. Based on the partition adaptation relationship, the mechanical parameters and the corresponding chloride ion parameters are weighted according to the preset weights, and the erosion influence coefficient is simultaneously introduced to achieve quantitative coupling to obtain the partition fusion parameters. Among them, the mechanical parameters include the damage factor of the tensile zone and the damage evolution parameter of the compressive zone, and the chloride ion parameters include the penetration depth and concentration gradient. The partition fusion parameters are spatially mapped according to the penetration depth, and gradient parameters that continuously change along the erosion depth are formed through interpolation to fit the actual spatial distribution of degradation; Verify the gradient parameters based on the uniaxial tension / compression constitutive equation of concrete, and correct the parameters that deviate from the physical laws according to the constraints; The corrected gradient parameters are associated with the tension / compression partition identification and penetration depth coordinates to form comprehensive material parameters that include mechanical degradation, chloride ion corrosion and spatial distribution characteristics.

9. A concrete degradation simulation system under continuous load-chloride ion coupling, characterized in that: The invention comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a method for simulating concrete deterioration under continuous load-chloride ion coupling as claimed in any one of claims 1 to 8.

Images