A digital twin crack prevention and control method and system for a thin-walled mass concrete pool

Abstract

The present application relates to the technical field of concrete structure crack control, in particular to a thin-walled mass concrete pool digital twin crack prevention and control method and system, the prevention and control method comprises the following steps: constructing a finite element model considering the coupling of temperature field, stress field and shrinkage and creep; collecting temperature data, strain data and environmental parameter data of the concrete structure, dynamically updating and parameter calibrating the finite element model; dynamically predicting the temperature field, stress field and crack risk of the concrete structure; generating crack resistance strategies including temperature control and maintenance strategies and crack resistance control strategies; the prevention and control system automatically coordinates crack prevention and control measures according to the strategy instructions. The prevention and control system comprises a model construction module, a data acquisition module, a model dynamic calibration module, a multi-field coupling analysis module, a crack risk assessment module, a crack resistance strategy generation module, an intelligent temperature control and maintenance execution module, a central processing unit and a water storage test analysis module.

Description

Technical Field

[0001] This invention relates to the field of concrete crack prevention and control technology, specifically to a digital twin crack prevention and control method and system for thin-walled, large-volume concrete water tanks. Background Technology

[0002] With the rapid development of urban water supply and drainage projects, sewage treatment projects, and underground water storage projects, large reinforced concrete water tanks have been widely used. Concrete water tanks are typically characterized by large structural dimensions, high requirements for seepage prevention, and long construction periods. Their tank walls are susceptible to cracks due to temperature changes, shrinkage deformation, and structural constraints during construction and operation.

[0003] After the concrete is poured, the cement hydration reaction continues to release heat, causing the internal temperature of the concrete to rise continuously. Meanwhile, the surface of the pool wall dissipates heat quickly upon contact with the air, resulting in a significant temperature difference between the inside and outside. When the temperature stress exceeds the tensile strength of the concrete at the same time, temperature cracks are likely to form in the pool wall.

[0004] Meanwhile, concrete primarily undergoes drying shrinkage during the hardening process. This is especially true for thin-walled concrete pools, where the exposed surface area is large, leading to faster moisture loss and more pronounced shrinkage deformation. When this shrinkage deformation is constrained by the base slab, adjacent pool walls, and construction joints, it further generates tensile shrinkage stress, increasing the risk of cracking.

[0005] In addition, concrete also exhibits creep during long-term stress. Creep can release temperature and shrinkage stresses to some extent, but it also leads to the continuous development of structural deformation. Therefore, the impact of creep on the redistribution of structural stress needs to be comprehensively considered in crack risk analysis.

[0006] Compared to ordinary large-volume concrete structures, concrete water tank walls have more complex structural characteristics. The walls are relatively thin, but long and high, making them typical thin-walled large-volume structures. Both the inner and outer sides of the walls are directly exposed to the air, resulting in a double-sided heat dissipation structure. Furthermore, the surface moisture evaporates quickly, leading to significant time-varying shrinkage deformation. Traditional analytical methods struggle to accurately simulate the development of the temperature field, shrinkage deformation, and creep effects of the water tank walls simultaneously.

[0007] In actual construction, the formwork removal stage is one of the stages most prone to cracking. After the formwork is removed, the heat dissipation conditions and surface water loss of the pool wall will change significantly, easily leading to rapid changes in surface temperature and rapid development of shrinkage deformation, thus forming large temperature stress and shrinkage stress. However, existing analysis methods usually use fixed boundary conditions, which cannot reflect the impact of changes in boundary conditions before and after formwork removal on the temperature field and shrinkage deformation.

[0008] In addition, a water filling and impoundment test is required after the water tank construction is completed. During the impoundment process, the stress state of the tank wall changes, and some micro-cracks may further expand under water pressure, leading to leakage problems. Currently, most crack prevention technologies only focus on temperature cracks during the construction phase, lacking the ability to analyze the entire process of shrinkage cracks, creep deformation, and crack expansion during the impoundment phase.

[0009] In recent years, digital twin technology has been increasingly applied to the field of structural health monitoring in engineering. Digital twins establish virtual models that operate synchronously with the physical structure, enabling real-time mapping and dynamic analysis of structural conditions. However, existing digital twin technologies are mostly applied to bridge and building structures, and research on the collaborative prevention and control of cracks throughout the entire process of concrete water tanks is still limited, especially lacking a digital twin closed-loop management system that simultaneously considers the coupled effects of temperature, shrinkage, and creep.

[0010] Therefore, it is necessary to propose a digital twin-based whole-process crack collaborative prevention and control method suitable for the structural characteristics of concrete water tanks, so as to realize intelligent health management of concrete water tanks throughout their entire life cycle. Summary of the Invention

[0011] The technical problem to be solved by this invention is to provide a digital twin crack prevention and control method and system for thin-walled large-volume concrete water tanks, establish a digital twin closed-loop management system covering the entire construction process, and comprehensively consider the coupling effect of temperature field, shrinkage deformation and creep effect in the whole process analysis to realize real-time perception, dynamic analysis, intelligent early warning and active control of crack risk in thin-walled large-volume concrete water tanks throughout the entire life cycle.

[0012] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0013] A digital twin crack prevention method for thin-walled, large-volume concrete water tanks includes:

[0014] S1. By integrating concrete mix proportion parameters, concrete structure geometric parameters, and environmental parameters, a digital twin finite element model of a thin-walled, large-volume concrete water tank is constructed, which considers the coupling effects of temperature field, stress field, and shrinkage and creep. The finite element model is used to simulate the hydration heat release and temperature conduction process of the concrete structure, as well as the development process of temperature stress and shrinkage and creep stress.

[0015] S2. During the construction phase, temperature data, strain data, and environmental parameter data of the concrete structure are automatically collected, and the finite element model is dynamically updated and its parameters are calibrated.

[0016] S3. Dynamic prediction of temperature field, stress field and crack risk of concrete structure through digital twin through finite element model;

[0017] S4. Based on the dynamic prediction results of crack risk, combined with the analysis results of safety factor, peak temperature, temperature difference between inside and outside and shrinkage stress, generate crack resistance strategies including temperature control maintenance strategy and crack resistance control strategy;

[0018] S5. The prevention and control system automatically coordinates crack prevention and control measures according to strategy instructions.

[0019] As a further optimization of the present invention, the construction of the digital twin finite element model in S1 includes:

[0020] S11. Obtain concrete mix proportion parameters, concrete structure geometric parameters, and environmental parameters;

[0021] S12. Based on the geometric parameters of the concrete structure, a three-dimensional finite element model is established and meshed using finite element analysis software;

[0022] S13. Obtain time-varying parameters of the thermal and mechanical properties of materials through experiments, and import the material property parameters into the simulation environment through a software interface;

[0023] S14. Embed subroutines for temperature field analysis, shrinkage analysis, and creep analysis through the secondary development interface of finite element software to perform temperature field simulation and stress field simulation;

[0024] S15. Based on environmental parameters, generate thermal boundary conditions for the finite element model through a subroutine;

[0025] S16. Set the initial conditions and structural constraints of the model, configure the coupled calculation logic of temperature field, stress field and shrinkage and creep, and complete the construction of the coupled finite element model.

[0026] As a further optimization of the present invention, in S11, the concrete mix proportion parameters include cement type, water-cement ratio, aggregate gradation and admixture dosage parameters.

[0027] The geometric parameters of concrete structures include the three-dimensional dimensions, spatial shape, and boundary constraints of the components.

[0028] Environmental parameters include ambient temperature, relative humidity, and wind speed.

[0029] As a further optimization of the present invention, in S14, the temperature field simulation adopts the three-dimensional unsteady heat conduction theory; during the analysis, the concrete hydration heat release model is written using the ABAQUS user subroutine UMATHT, and the heat release law of hydration heat release with age is loaded into the finite element calculation process.

[0030] In S15, the ambient temperature and convective heat transfer coefficient are defined using the FILM user subroutine, and the thermal boundary conditions are generated.

[0031] As a further optimization of the present invention, when configuring the coupled calculation logic of temperature field, stress field and shrinkage and creep in S16, the material constitutive model adopts an elastic model that considers the influence of age, and the coupled calculation logic includes:

[0032] Temperature field simulation: Dynamic input of thermal boundary conditions is achieved through the ABAQUS secondary development interface; and the transient thermal analysis module is used to perform numerical calculations on the entire process of heat release during concrete hydration.

[0033] Stress field simulation: The temperature strain, shrinkage strain and creep strain are uniformly introduced into the finite element analysis process through the UEXPAN subroutine, and the elastic modulus is dynamically updated with age by combining the UFIELD subroutine, thus establishing a coupled analysis model of the entire process of temperature, shrinkage and creep.

[0034] The stress field simulation is achieved by using a sequential coupling analysis method based on the temperature field calculation results. The temperature field results obtained at each time step are imported into the structural analysis module as load conditions, and the coupled analysis model is used to calculate the stress and deformation development process of the concrete structure under temperature action.

[0035] Furthermore, this embodiment employs a time-step-based sequential coupling calculation method for the temperature field and structural field. At any calculation time step (t... k Within the central processing unit, the current concrete age and the corresponding ambient temperature, relative humidity, wind speed, and site boundary status information are first read.

[0036] During the temperature field analysis phase, the UMATHT user subroutine is invoked to determine the concrete hydration heat release state corresponding to the current time step based on the concrete age and a pre-established hydration heat release model. Simultaneously, the FILM user subroutine is invoked to determine the convective heat transfer conditions on the outer surface of the structure based on the ambient temperature, wind speed, and the concrete surface coverage. The finite element solver then performs the three-dimensional unsteady-state heat conduction calculation for the current time step, obtaining the temperature values ​​of all finite element nodes and the temperature change relative to the previous time step.

[0037] After completing the temperature field calculation, the nodal temperature field obtained at the current time step is passed to the structural analysis module as a predefined field. During the structural analysis phase, the UEXPAN user subroutine is called to calculate the strain increment caused by temperature change, concrete shrinkage development, and creep development based on the current age and temperature change, and the strain increment is written to the corresponding finite element integration point; at the same time, the UFIELD user subroutine is called to update the material field variables according to the current concrete age, and obtain the elastic modulus corresponding to the current age according to the pre-established elastic modulus age development model.

[0038] The structural solver further calculates the principal stresses, equivalent stresses, and structural deformations of the concrete, taking into account the current structural constraints. After the current time step calculation is completed, the nodal temperatures, material state variables, strain states, and stress states are saved and used as the basis for the next time step (t). {k+1} The initial state of the concrete is determined by the calculation process described above, which proceeds step by step until the entire process analysis from concrete pouring, heating, peak temperature, cooling to demolding and curing is completed. This enables the coupled calculation of temperature, shrinkage and creep as they develop over time.

[0039] As a further optimization of the present invention, the dynamic updating and parameter calibration of the finite element model in S2 includes:

[0040] S21. The collected temperature and strain data are sent to the central processing unit, and a mapping relationship is established between the real-time monitoring data and the corresponding nodes or elements of the finite element model according to the spatial location of the sensor. S22. The central processing unit adopts the weighted least squares data assimilation method to compare the real-time temperature and strain monitoring data with the finite element model prediction data at the corresponding locations, establishes the model calibration objective function, and calculates the relative error of temperature prediction between the finite element model predicted temperature and the actual measured temperature, as well as the relative error of strain prediction between the finite element model predicted strain and the actual measured strain. S23. When the relative error of the temperature prediction is greater than 2% or the relative error of the strain prediction is greater than 5%, the parameter optimization algorithm is used to iteratively correct the hydration exothermic model parameters, thermal boundary condition parameters and material performance parameters; S24. Recalculate the temperature and stress fields for the current time step using the corrected model parameters, and recalculate the relative error of temperature prediction and the relative error of strain prediction. When the relative error of temperature prediction is no greater than 2% and the relative error of strain prediction is no greater than 5%, the current finite element model is determined to meet the preset calibration requirements. The corrected model parameters are saved and used as the model parameters for subsequent time steps to achieve synchronous updates between the digital twin finite element model and the actual concrete structure state. If any prediction relative error still exceeds the corresponding preset calibration allowable error, parameter iterative correction continues until the preset calibration allowable error is met or the preset maximum number of iterations is reached.

[0041] As a further optimization of the present invention, the temperature strain is expressed as:

[0042] ;

[0043] Where α is the coefficient of linear expansion of concrete, and ΔT is the temperature change;

[0044] Contraction strain is expressed as:

[0045] ;

[0046] Where t represents the concrete age considered in the calculation; t s The concrete age at the onset of shrinkage; The age at the onset of contraction is t. s The calculation considers the shrinkage strain at an age of t; β is the nominal contraction coefficient; β is the coefficient of contraction over time.

[0047] Creep strain is expressed as:

[0048] ;

[0049] in, For creep response, σ is the creep coefficient, σ is the current stress, and E(t) is the current elastic modulus.

[0050] As a further optimization of the present invention, the full-process digital twin dynamic prediction of crack risk in S3 includes:

[0051] S31. Obtain the tensile strength of concrete and establish a tensile strength growth model based on the test results at different ages;

[0052] S32. Set appropriate safety factors according to the importance of the concrete structure, and use the ratio of tensile strength to safety factor as the threshold for judging crack risk;

[0053] S33. Based on the stress field results calculated by the finite element model, extract the tensile stress distribution and automatically extract and mark the finite element elements whose tensile stress exceeds the threshold.

[0054] S34. Combining stress isosurface analysis and regional clustering algorithms, continuous high-stress regions are identified and statistically analyzed to obtain the spatial location, range, and maximum stress value of each continuous high-stress region, and crack risk levels are classified.

[0055] As a further optimization of the present invention, it also includes water retention tests, crack repair and re-inspection after concrete structure demolding and curing, specifically including:

[0056] S6. Based on the finite element model, a fluid-structure interaction analysis model is further established. The water storage load is applied in stages to the pool wall and bottom plate structure as an external force to simulate the development process of structural stress and deformation under water pressure. Temperature data, strain data and environmental parameter data of the concrete structure are automatically collected to dynamically update and calibrate the parameters of the finite element model. The collected data on pool wall displacement, strain and crack changes are simultaneously fed back to the digital twin platform in real time.

[0057] S7. During the water storage monitoring process, based on the comparison and analysis of real-time collected data and finite element prediction results, high-risk areas in the water storage stage are automatically identified and crack early warning information is generated; the crack propagation risk can be predicted by combining the water pressure change trend and crack propagation direction.

[0058] S8. Automatically generate corresponding repair plans based on crack type, crack width, and leakage level; automatically generate grouting path, grouting pressure, and grouting duration according to crack location, and execute repair operations through intelligent grouting equipment;

[0059] S9. After the crack repair is completed, a re-inspection and secondary water storage verification stage will be carried out.

[0060] A digital twin crack prevention and control system for thin-walled, large-volume concrete water tanks is also provided to implement the aforementioned prevention and control method, comprising:

[0061] The model building module is used to build digital twin finite element models of thin-walled, large-volume concrete structures.

[0062] The data acquisition module includes temperature sensors and strain gauges embedded in the concrete structure, as well as temperature and humidity monitoring devices and environmental acquisition devices deployed on the concrete surface.

[0063] The model dynamic calibration module compares, analyzes, and corrects the real-time monitoring data acquired by the data acquisition module with the predicted values ​​of the finite element model.

[0064] A multi-field coupling analysis module, wherein the multi-field coupling analysis module embeds temperature field analysis subroutines, shrinkage analysis subroutines and creep analysis subroutines through the secondary development interface of finite element software;

[0065] The crack risk assessment module automatically identifies and dynamically assesses high-risk crack areas based on stress analysis results and concrete strength development patterns.

[0066] The crack resistance strategy generation module generates temperature control maintenance strategies and crack resistance control strategies based on the prediction results of the crack risk assessment module.

[0067] The intelligent temperature control curing execution module includes a coordinated cooling system, a spraying system, and a thermal insulation covering system. Based on the strategy generated by the anti-cracking strategy generation module, it automatically coordinates and executes thermal insulation, spraying, cooling, and demolding sequence.

[0068] The central processing unit is used to receive and process real-time monitoring data transmitted by the data acquisition module, coordinate the operation of each module, execute data assimilation algorithms and parameter optimization algorithms, generate control commands and distribute them to each execution module.

[0069] The beneficial effects of adopting the above technical solution are as follows:

[0070] The prevention and control method provided by this invention constructs a digital twin finite element model of a thin-walled, large-volume concrete water tank that considers the coupling effects of temperature field, stress field, and shrinkage and creep. The finite element model simulates the hydration heat release and temperature conduction process of the concrete structure, as well as the stress evolution law caused by the combined effects of temperature stress, volume shrinkage, and creep. Combined with monitoring data from the construction and water injection stages, the method dynamically predicts the risk of cracks throughout the entire structural process. This achieves proactive, precise, and comprehensive collaborative prevention and control of cracks in thin-walled, large-volume concrete based on digital twin simulation. It can significantly reduce temperature cracks, shrinkage cracks, and leakage cracks in thin-walled, large-volume concrete water tanks during construction and early service, thereby improving structural integrity, impermeability, and service life, and reducing later maintenance costs.

[0071] The aforementioned prevention and control system achieves closed-loop operation throughout the entire process, multi-field coupling, virtual-real synchronization, intelligent early warning, and proactive control, realizing a leap from passive repair to proactive prevention and control, and from local temperature control to collaborative management of cracks throughout the entire life cycle. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of the present invention clearer, the invention will be described clearly and completely below in conjunction with specific embodiments.

[0073] A digital twin crack prevention method for thin-walled, large-volume concrete water tanks includes:

[0074] S1. By integrating concrete mix proportion parameters, concrete structural geometric parameters, and environmental parameters, a digital twin finite element model of a thin-walled, large-volume concrete water tank is constructed, considering the coupled effects of temperature field, stress field, and shrinkage / creep. This finite element model simulates the hydration heat release and temperature conduction processes of the concrete structure, as well as the stress evolution caused by the combined effects of temperature stress, volume shrinkage, and creep. Environmental parameters include ambient temperature, relative humidity, and wind speed. Construction conditions and structural boundary conditions are set according to the actual construction state of the concrete and used as boundary condition inputs for the finite element model.

[0075] S2. During the construction phase, temperature data, strain data and environmental parameter data of the concrete structure are automatically collected by temperature sensors, strain gauges and environmental monitoring equipment, and the finite element model is dynamically updated and the parameters are calibrated.

[0076] S3. By analyzing the internal temperature gradient, constraint conditions, and stress distribution characteristics of concrete using a finite element model, high-risk areas of temperature stress and shrinkage stress concentration in concrete structures are identified. A full-process digital twin dynamic prediction of the temperature field, stress field, and crack risk of concrete structures is conducted. Specifically, based on the development of concrete tensile strength, the probability of crack occurrence and crack-sensitive areas are predicted, providing a basis for the formulation of subsequent temperature control curing strategies and crack resistance control strategies.

[0077] S4. Based on the dynamic prediction results of crack risk, combined with the analysis results of safety factor, peak temperature, temperature difference between inside and outside and shrinkage stress, generate crack resistance strategies including temperature control maintenance strategy and crack resistance control strategy;

[0078] S5. The prevention and control system automatically coordinates crack prevention and control measures according to strategy instructions.

[0079] A digital twin finite element model is a high-fidelity virtual mapping system based on computer simulation. By integrating concrete mix proportion parameters, concrete structure geometric parameters, and environmental parameters, it aims to construct a digital model that evolves synchronously with the actual engineering project. In this embodiment, constructing the digital twin finite element model in S1 includes:

[0080] S11. Obtain concrete mix proportion parameters, concrete structure geometric parameters, and environmental parameters;

[0081] S12. Based on the geometric parameters of the concrete structure, a three-dimensional finite element model is established and meshed using finite element analysis software to form a three-dimensional mesh model of the concrete;

[0082] S13. Obtain relevant material performance parameters, including time-varying parameters of material thermal and mechanical properties, through experiments, specifically including thermal conductivity, specific heat capacity, elastic modulus, etc.; import the material performance parameters into the simulation environment through software interface;

[0083] S14. Embed subroutines for temperature field analysis, shrinkage analysis, and creep analysis through the secondary development interface of finite element software to simulate the release of heat of hydration, temperature transfer, and stress development process of concrete.

[0084] S15. Based on environmental parameters, generate thermal boundary conditions for the finite element model through a subroutine;

[0085] S16. Set the initial conditions and structural constraints of the model, configure the coupled calculation logic of temperature field, stress field and shrinkage and creep, and complete the construction of the coupled finite element model.

[0086] In some feasible examples, in S11, the concrete mix proportion parameters include cement type, water-cement ratio, aggregate gradation and admixture dosage parameters, which can be obtained based on laboratory mix design reports and construction quality test data;

[0087] The geometric parameters of concrete structures include the three-dimensional dimensions, spatial shape, and boundary constraints of the components; parametric data can be exported from building information models, or point cloud information can be obtained by three-dimensional laser scanning and then reconstructed to form a computational model;

[0088] Environmental parameters, including ambient temperature, relative humidity, and wind speed, can be acquired in real time through temperature and humidity monitoring equipment and environmental acquisition devices deployed at the construction site, and the data acquisition devices will periodically transmit the data to the central processing unit.

[0089] In some feasible examples, such as S14, the temperature field simulation adopts the three-dimensional unsteady heat conduction theory; during the analysis, the ABAQUS user subroutine UMATHT is used to write a concrete hydration heat release model, and the heat release law of hydration heat release with age is loaded into the finite element calculation process.

[0090] In S15, the ambient temperature and convective heat transfer coefficient are defined using the FILM user subroutine to generate the thermal boundary conditions.

[0091] In some feasible examples, when configuring the coupled calculation logic for temperature field, stress field, and shrinkage / creep in S16, the material constitutive model adopts an elastic model that considers the effects of age. The coupled calculation logic includes:

[0092] Temperature field simulation: Dynamic input of thermal boundary conditions is achieved through the ABAQUS secondary development interface; and the transient thermal analysis module is used to perform numerical calculations on the entire process of heat release during concrete hydration.

[0093] Stress field simulation: The temperature strain, shrinkage strain and creep strain are uniformly introduced into the finite element analysis process through the UEXPAN subroutine, and the elastic modulus is dynamically updated with age by combining the UFIELD subroutine, thus establishing a coupled analysis model of the entire process of temperature, shrinkage and creep.

[0094] The stress field simulation is achieved by using a sequential coupling analysis method based on the temperature field calculation results. The temperature field results obtained at each time step are imported into the structural analysis module as load conditions, and the coupled analysis model is used to calculate the stress and deformation development process of the concrete structure under temperature action.

[0095] Specifically, when configuring the coupled calculation logic for temperature field, stress field, and shrinkage / creep in S16, a sequential coupling analysis method based on time steps is adopted, which includes:

[0096] Within the k-th calculation time step, obtain the current concrete age and environmental parameters;

[0097] Call the UMATHT user subroutine to determine the corresponding hydration heat release state based on the current concrete age and hydration heat release model, and introduce it into the transient temperature field calculation process.

[0098] The FILM user subroutine is called to update the convective heat transfer boundary conditions on the surface of the concrete structure based on the current ambient temperature and convective heat transfer coefficient. The temperature field corresponding to the k-th calculation time step is obtained through transient thermal analysis.

[0099] The temperature field calculation results are transferred to the structural analysis module. The strain increments corresponding to temperature changes, shrinkage development and creep development are introduced into the structural analysis process through the UEXPAN user subroutine. The elastic modulus is updated according to the current age of the concrete through the UFIELD user subroutine.

[0100] The stress and deformation fields at the current time step are calculated based on the updated strain state, material property parameters, and structural constraints.

[0101] The temperature field, stress field, material state parameters, and strain state obtained at the current time step are used as the initial state for the next calculation time step. The above process is repeated until the coupled analysis of the entire process of temperature, shrinkage, and creep within the set analysis period is completed.

[0102] In some feasible examples, concrete structures undergo thermal expansion or contraction due to the exothermic effects of hydration and changes in ambient temperature. Expansion deformation occurs when the internal temperature of the concrete rises, while contraction deformation occurs during the cooling phase. Under structural constraints, this temperature deformation further generates thermal stress, which is one of the important causes of early cracking in large-volume concrete. Temperature strain can be expressed as:

[0103] ;

[0104] Where α is the coefficient of linear expansion of concrete, and ΔT is the temperature change;

[0105] During the hardening process of concrete structures, drying shrinkage mainly occurs due to internal moisture loss and continuous cement hydration. Its shrinkage strain can be expressed as:

[0106] ;

[0107] Where t represents the concrete age considered in the calculation; t s The concrete age at the onset of shrinkage; The age at the onset of contraction is t. s The calculation considers the shrinkage strain at an age of t; β is the nominal contraction coefficient; β is the coefficient of contraction over time.

[0108] Creep causes the internal confinement stress of concrete structures to gradually relax over time, thus significantly influencing the development of temperature and shrinkage cracks. In numerical analysis, the creep model is coupled with the shrinkage and temperature field models to simulate the time-varying behavior of concrete throughout its early stages. The creep strain expression can be given as:

[0109] ;

[0110] in, For creep response, σ is the creep coefficient, σ is the current stress, and E(t) is the current elastic modulus.

[0111] In finite element analysis, temperature strain, shrinkage strain, and creep strain are comprehensively considered, and the growth law of elastic modulus with age is combined. Through UEXPAN, UFIELD and related subroutines, multi-field coupled analysis of temperature, shrinkage and creep of water tank structure is realized, so as to obtain the stress evolution and crack development law of concrete structure throughout the process.

[0112] In some feasible examples, the dynamic updating and parameter calibration of the finite element model in S2 are implemented based on real-time monitoring data, specifically including:

[0113] S21. The collected temperature and strain data are sent to the central processing unit, and a mapping relationship is established between the real-time monitoring data and the corresponding nodes or elements of the finite element model according to the spatial location of the sensor. S22. The central processing unit adopts the weighted least squares data assimilation method to compare the real-time temperature and strain monitoring data with the finite element model prediction data at the corresponding locations, establishes the model calibration objective function, and calculates the relative error of temperature prediction between the finite element model predicted temperature and the actual measured temperature, as well as the relative error of strain prediction between the finite element model predicted strain and the actual measured strain. S23. When the relative error of the temperature prediction is greater than 2% or the relative error of the strain prediction is greater than 5%, the parameter optimization algorithm is used to iteratively correct the hydration exothermic model parameters, thermal boundary condition parameters and material performance parameters; S24. Recalculate the temperature and stress fields for the current time step using the corrected model parameters, and recalculate the relative error of temperature prediction and the relative error of strain prediction. When the relative error of temperature prediction is no greater than 2% and the relative error of strain prediction is no greater than 5%, the current finite element model is determined to meet the preset calibration requirements. The corrected model parameters are saved and used as the model parameters for subsequent time steps to achieve synchronous updates between the digital twin finite element model and the actual concrete structure state. If any prediction relative error still exceeds the corresponding preset calibration allowable error, parameter iterative correction continues until the preset calibration allowable error is met or the preset maximum number of iterations is reached.

[0114] Furthermore, the model dynamic calibration uses real-time temperature and strain data collected on-site as observations. Assuming there are m temperature monitoring points and n strain monitoring points in the current calculation time step, the nodes or elements corresponding to the positions of each sensor are determined in the finite element model based on the spatial coordinates of each sensor in the concrete structure, and the measured temperature, finite element predicted temperature, measured strain, and finite element predicted strain are obtained respectively.

[0115] The central processing unit uses a weighted least squares method to assimilate real-time monitoring data and finite element model prediction data, and establishes a model calibration objective function:

[0116]

[0117] Where θ represents the set of parameters of the finite element model to be calibrated; T i obs and T i cal ε represents the measured temperature and the finite element predicted temperature at the i-th temperature monitoring point, respectively; j obs and ε j cal Let w represent the measured strain and the finite element predicted strain at the j-th strain monitoring point, respectively; T,i and w ε,j These represent the weights corresponding to temperature data and strain data, respectively.

[0118] For monitoring sensors of the same type and accuracy level that are in normal working condition, the weights corresponding to temperature data and strain data are set equally. When there are differences in the accuracy of sensors at different monitoring points, the corresponding weights are determined according to the sensor calibration accuracy, and monitoring data with higher measurement accuracy are given higher weights.

[0119] The set of parameters to be calibrated includes one or more of the following: hydration exothermic model parameters, equivalent thermal conductivity, surface convective heat transfer coefficient, and elastic modulus correction parameters. Temperature monitoring data is mainly used to calibrate the hydration exothermic model parameters, equivalent thermal conductivity, and surface convective heat transfer coefficient, while strain monitoring data is mainly used to calibrate the elastic modulus and related mechanical property parameters.

[0120] Specifically, the preset calibration allowable errors include temperature calibration allowable errors and strain calibration allowable errors. The relative error between the temperature predicted by the finite element model and the actual measured temperature is no greater than 2%, and the relative error between the strain predicted by the finite element model and the strain measured on-site is no greater than 5%. The weighted least squares method is used to optimize the parameters of the model to be calibrated by comprehensively considering the deviation between the real-time monitoring data and the predicted data from the finite element model. The preset calibration allowable errors are used to determine whether the parameter calibration results of the finite element model meet the requirements.

[0121] When the relative error between the temperature predicted by the finite element model and the actual temperature measured on site is greater than 2%, or the relative error between the strain predicted by the finite element model and the strain measured on site is greater than 5%, the central processing unit calls the parameter optimization program to iteratively correct the parameters to be calibrated based on the weighted least squares model calibration results. The updated model parameters are then used to re-execute the temperature field and stress field calculations for the current time step, and the finite element model prediction results are compared and analyzed with the actual measured data again.

[0122] The iterative process of "real-time data acquisition - model prediction - error analysis - parameter correction - model recalculation" is repeated. When the relative error between the temperature predicted by the finite element model and the actual measured temperature is no greater than 2%, and the relative error between the strain predicted by the finite element model and the actual measured strain is no greater than 5%, the current finite element model is determined to have met the preset calibration requirements. The parameter iteration calibration at the current time step is stopped, and the corrected model parameters are saved and transferred to the next calculation time step. If any relative error still exceeds the corresponding preset calibration allowable error, the parameter iteration correction continues until the preset calibration allowable error is met or the preset maximum number of iterations is reached. This achieves dynamic updating of the digital twin finite element model according to the actual state of the concrete structure.

[0123] In some feasible examples, dynamic prediction of crack risk in S3 via a digital twin throughout the entire process includes:

[0124] S31. Obtain the tensile strength of concrete and establish a tensile strength growth model based on the test results at different ages;

[0125] S32. Set a corresponding safety factor according to the importance of the concrete structure, and use the ratio of tensile strength to safety factor as the allowable tensile stress at age t, and use this as the threshold for judging crack risk.

[0126] S33. Based on the stress field results calculated using the finite element model, extract the maximum principal tensile stress σ of the i-th finite element. 1,i The ratio of the maximum principal tensile stress to the allowable tensile stress is used as the crack risk coefficient of the i-th finite element.

[0127] S34. Based on the crack risk coefficient, risk identification is performed on finite element elements, and based on the topological adjacency relationship between finite element mesh elements, connectivity retrieval is performed on adjacent risk elements; adjacent risk elements with common surfaces are divided into the same continuous risk region, and the spatial location, range, maximum principal tensile stress and maximum crack risk coefficient of each continuous risk region are statistically analyzed.

[0128] S35. Crack risk levels are classified according to crack risk coefficients. When the coefficient is <0.7, it is classified as low risk; when 0.7≤coefficient<1.0, it is classified as medium risk; when the coefficient is ≥1.0, it is classified as high risk. For a continuous risk area composed of multiple finite element elements, the largest crack risk coefficient in the area is used as the basis for determining the risk level of the area.

[0129] Specific implementation process:

[0130] Simulating the development process of internal temperature and stress in concrete structures based on the coupling relationship between temperature field, stress field, and shrinkage and creep is the core analytical step of this crack prevention method. Its essence lies in reproducing the coupling mechanism of temperature field and stress field during the hydration process of concrete structure through numerical simulation technology, thereby identifying potential high-risk areas for cracks.

[0131] Specifically, the simulation process begins by establishing an engineering monitoring and data acquisition system. Strain gauges with temperature monitoring capabilities are pre-embedded at key locations within the concrete structure, such as the middle and corners of the pool wall and areas with strong constraints. These monitoring points are arranged according to a preset grid spacing. Each strain gauge is connected to a central processing unit via a data acquisition device and automatically collects internal temperature and strain data of the concrete structure at set time intervals, which are then transmitted to the server in real time via a wireless communication network. Simultaneously, temperature and humidity monitoring devices are deployed on the surface of the concrete structure to acquire boundary environmental parameters such as surface temperature, ambient temperature, and relative humidity, providing real-time boundary conditions for the temperature field simulation.

[0132] Secondly, based on the acquired real-time data, a finite element model considering factors such as hydration heat release and temperature conduction is established in the finite element simulation platform. The finite element model simulates the hydration heat release and heat dissipation processes of the concrete structure to obtain the temperature field distribution at different ages, and further calculates the stress development process caused by temperature gradient, drying shrinkage, and external constraints. Subsequently, combined with the growth law of concrete tensile strength with age, areas of concentrated tensile stress inside the concrete structure are identified; when the local tensile stress exceeds the safety threshold for the corresponding age, the area is determined to have a high risk of cracking. Furthermore, by comparing the real-time monitoring data with the numerical model prediction results, the model parameters can be dynamically corrected, specifically including thermal boundary conditions and thermal conductivity, thereby improving the accuracy of temperature and stress field simulation and achieving synchronous evolution between the digital model and the actual structural state, providing a reliable basis for subsequent crack risk prediction and assessment and crack resistance strategies.

[0133] The coupled relationship between temperature field, stress field, and shrinkage / creep reflects the interaction mechanism between heat transfer, volume deformation, and mechanical response within concrete structures. The key to its numerical simulation lies in establishing a relatively accurate hydration heat release model, thermal parameter model, and time-varying mechanical property model. First, based on the cement type, cement content, and mineral admixture dosage in the concrete mix design, isothermal calorimetry tests were conducted to obtain the hydration heat release characteristics at different ages. During the experiments, a multi-channel isothermal calorimeter continuously recorded the hydration heat release power of concrete at early ages and automatically collected heat flow data at preset time intervals. By fitting the hydration heat release curves to a function, a concrete hydration heat release model was established to describe the changes in the hydration heat release rate and cumulative heat release at different ages. In determining the thermal parameters, a thermal parameter model was established based on the volume ratio of the various components of the concrete. The thermal conductivity, specific heat capacity, and density of cement paste, coarse and fine aggregates, and the interface transition zone can be determined through experimental testing. Then, the weighted average method is used to calculate the overall equivalent thermal conductivity and equivalent specific heat capacity of concrete, and the relevant parameters are input into the finite element model to simulate the heat transfer process and temperature distribution law inside the concrete.

[0134] The thermal conductivity of concrete was determined as follows: A cylindrical specimen with a diameter of 200 mm and a height of 400 mm was used. During molding, a 40 mm diameter iron rod was pre-embedded in the center to form a through hole. After curing, the specimen was installed in a thermal conductivity meter. Water was filled into the inner cavity of the specimen, and a heater was inserted. The external cooling tank was circulated with water to maintain a constant water level. The heater was turned on to raise the water temperature in the inner cavity, and the water was stirred to ensure uniform temperature. After the voltage, current, and center temperature stabilized, the cooling water temperature and center water temperature were measured every 10 minutes. The voltage and current were recorded, the heating power (W) was calculated, and the thermal conductivity was calculated according to the formula. The average of all measurements was taken as the result for the specimen.

[0135] The specific heat capacity of concrete is determined as follows: A cylindrical specimen with a diameter of 200 mm and a height of 400 mm is used, with a 40 mm diameter iron rod pre-embedded in the center during molding to form a through hole. After curing, the specimen is wiped dry and weighed, then placed in an adiabatic heater specimen container and a fixed amount of water is added. After the system temperature stabilizes, the initial temperature and meter reading are recorded. The water temperature is then heated by electricity until it rises by 10-15°C, and then stopped. After the temperature becomes uniform, the final temperature and power consumption are recorded. This process is repeated three times. Based on the heat supplied by the heater and the agitator, the heat absorbed by the specimen container and water is deducted, and the heat Q absorbed by the specimen is calculated. Then, combined with the specimen mass M and the initial and final temperatures of the three heating cycles, the coefficients of the quadratic expression for the specific heat capacity as a function of temperature are solved simultaneously to obtain the specific heat capacity c. The average value of two specimens is taken.

[0136] In the process of determining thermal parameters, a time-varying thermal parameter model is established based on the volume ratio of each component material in concrete. The relevant parameters are then input into the finite element model to simulate the heat transfer process and temperature distribution law inside the concrete.

[0137] Based on the aforementioned hydration heat release model and thermal parameter model, the temperature field simulation employs three-dimensional unsteady-state heat conduction theory. A three-dimensional geometric model consistent with the actual structure is established using the finite element software ABAQUS, and hexahedral heat conduction elements are used for mesh generation. For areas with significant temperature gradient changes, such as edges, corners, and areas with concentrated constraints, local mesh refinement is performed to improve the accuracy and numerical stability of the temperature field calculation. The model boundary conditions comprehensively consider the heat exchange between the concrete structure and the external environment. Specifically, the influence of convective and radiative heat transfer on the surface of the concrete structure is considered simultaneously. The convective heat transfer coefficient is determined based on the measured wind speed, environmental conditions, and empirical formulas; the radiative heat transfer parameter is calculated based on the difference between the surface temperature of the concrete structure and the ambient temperature. For the formwork-covered area, the insulation-covered area, and the concrete structure contact interface, adiabatic boundary conditions or third-type boundary conditions are set according to the actual construction conditions to more realistically reflect the heat dissipation process and temperature evolution law during the concrete construction stage. In the temperature field solution process, a concrete hydration heat release model was written using the ABAQUS user subroutine UMATHT, and the heat release law of hydration heat release with age was loaded into the finite element calculation process. Simultaneously, the FILM subroutine was used to define the ambient temperature and convective heat transfer coefficient, and the dynamic input of temperature boundary conditions was achieved through the ABAQUS secondary development interface. Subsequently, the transient thermal analysis module was used to numerically calculate the entire process of concrete hydration heat, and the finite element calculation time step was set according to the sampling period of the field temperature and strain monitoring data. The calculation time step was no greater than the minimum sampling period of the field monitoring data. The calculation time step was shortened during stages of rapid temperature rise, near the temperature peak, formwork removal, and changes in the insulation boundary to ensure the continuity of the temperature change process and the stability of the calculation.

[0138] During the calculation process, the temperature data obtained from real-time on-site monitoring is compared and analyzed with the calculation results of the finite element model. When the deviation between the model's predicted value and the measured value exceeds the preset calibration allowable error, the hydration heat parameters, thermal boundary conditions, and material thermal parameters are automatically corrected, realizing dynamic updating and iterative calibration of the model, thereby improving the simulation accuracy of the temperature and stress fields. Ultimately, the temperature field distribution law of the concrete structure at different ages and locations can be obtained, providing a theoretical basis for the temperature control and crack resistance design and construction optimization of large-volume concrete structures.

[0139] In the stress field simulation, the coupling effect between concrete temperature deformation, shrinkage deformation, and creep is further considered. Temperature deformation is determined by both the concrete's thermal expansion coefficient and temperature change; shrinkage deformation is mainly drying shrinkage; and creep reflects the additional deformation characteristics of concrete under continuous confined stress over time. The UEXPAN subroutine is used to uniformly incorporate temperature strain, shrinkage strain, and creep strain into the finite element analysis process, and the UFIELD subroutine is used to dynamically update the elastic modulus with age, thus establishing a coupled analysis model for the entire process of temperature, shrinkage, and creep.

[0140] The stress field simulation is achieved by using a sequential coupling analysis method based on the temperature field calculation results. That is, the temperature field results obtained at each time step are imported into the concrete structure analysis module as load conditions to calculate the stress and deformation development process of the concrete structure under temperature action.

[0141] The constitutive relation of concrete was studied using 150mm×150mm×300mm prism specimens as the object, and uniaxial compression tests were conducted on an electro-hydraulic servo high-rigidity pressure testing machine. Strain gauges, displacement sensors, or pre-embedded micro sensors were installed on the surface of the specimens to measure deformation. Force or displacement control was used during loading. After the peak stress, a stable stress-strain descent curve was obtained at a rate of 0.05mm / min through a closed-loop control system. At the same time, load-deformation data during the entire loading process were recorded. After data processing, the nominal stress and strain were converted into the true stress and strain. After removing discrete points, a mathematical model containing the ascending and descending segments was fitted. Based on multiple sets of repeated tests, key parameters (such as peak stress, peak strain, elastic modulus, and descending segment shape factor) were statistically calibrated, and the constitutive relation of the material was finally obtained. The development law of elastic modulus with age can be determined by concrete mechanics tests. These tests use standard prism specimens with side lengths of 150mm × 150mm × 300mm. Six specimens are prepared for each group, three for determining axial compressive strength and three for determining elastic modulus. The main instruments include a compression testing machine and micro-deformation measuring instruments (such as dial gauges and strain gauges), with a gauge length of 150mm. During the test, the axial compressive strength f is measured first. cpThen, place the other three specimens on the press, align them, and install deformation measuring instruments symmetrically on both sides of the specimens. Apply loads to the reference stress F0 and 1 / 3f. cp Deformation was recorded; if the difference in deformation between the two sides exceeded 20%, re-alignment was required. After two rounds of pre-loading, the elastic modulus was calculated, and the average of three specimens was taken. Poisson's ratio was determined based on the ratio of transverse strain to axial strain. Six 150mm×150mm×300mm prism specimens were used in each group. Using a press and a longitudinal and transverse deformation measuring instrument, the axial compressive strength was first measured, and then the other three specimens were loaded to 0.5MPa and 1 / 3 of the axial compressive strength. The longitudinal and transverse strains were recorded, and Poisson's ratio was calculated based on the ratio of the transverse strain increment to the longitudinal strain increment. The coefficient of thermal expansion was determined by a thermal deformation test. The test used a cylindrical specimen with a diameter of 200 mm and a height of 500 mm. A differential resistance strain gauge was vertically embedded in a sealed container. After molding and curing for 7 days, the specimen was placed in a constant temperature water tank. The resistance and temperature were measured after the initial temperature of 10-20℃ was maintained. The temperature was then raised to about 60℃ and maintained for a period of time before the final value was measured. The strain and temperature difference were calculated based on the change in resistance. The coefficient of linear expansion was obtained by formula, and the average value of the two specimens was taken.

[0142] Based on the above mechanical tests, a time-varying mechanical performance model of concrete structure was established. This model was then input into the finite element model. Combined with structural constraints, the stress field distribution of concrete under the combined action of temperature gradient, hydration shrinkage, and external constraints was calculated according to S16.

[0143] During the calculation process, the location of potential cracks in the concrete structure is identified by analyzing the principal tensile stress, equivalent stress, and stress concentration areas at different ages. The crack risk is then dynamically assessed based on the development law of concrete tensile strength. When the local tensile stress exceeds the safety threshold, the control system determines that the area has a high risk of cracking and feeds the analysis results back to the digital twin platform for subsequent temperature-controlled curing and crack-resistant measure optimization.

[0144] Boundary conditions are set to accurately reflect the constraint state in actual engineering. Specifically, the support positions and constraint forms can be determined based on structural design drawings and construction conditions. Fixed supports are subject to full constraint conditions, sliding supports are subject to tangential constraints based on contact friction characteristics, and elastic supports are simulated for their support stiffness using spring elements, thus realistically reflecting the stress state of the concrete structure during construction and service. The stress field solution uses a static structural analysis module, combined with a nonlinear iterative algorithm to calculate the structural response under temperature loads. During the calculation, convergence conditions such as force residuals and displacement increments are set, and the results of each time step are iteratively corrected to ensure the stability and accuracy of the numerical solution. Each time step can output parameters such as principal stress, shear stress, and equivalent stress, and a stress cloud map and stress extreme value distribution are generated through post-processing to analyze the stress development law inside the structure.

[0145] Crack risk identification is determined based on a comprehensive analysis of stress analysis results and the development law of concrete strength. The tensile strength of the material is determined by direct tensile testing using dog-bone (dumbbell-shaped) specimens on a servo-controlled testing machine at a loading rate of 0.1 mm / min. The fracture load is recorded and the tensile strength is calculated. Six specimens are tested at each age, and their strength characteristic values ​​are statistically analyzed. Simultaneously, a splitting test is used as an auxiliary verification method, using 150 mm × 150 mm × 150 mm cube specimens at a loading rate controlled at 0.05 MPa / s to supplement the evaluation of the material's tensile properties. Subsequently, a safety factor is set according to the structural importance, and the ratio of the structural tensile strength to the safety factor is used as the allowable tensile stress. After the finite element model completes the stress field calculation for the current time step, the maximum principal tensile stress of each finite element is extracted and divided by the corresponding allowable tensile stress to calculate the crack risk coefficient for the corresponding finite element: a coefficient < 0.7 indicates low risk; 0.7 ≤ coefficient < 1.0 indicates medium risk; and a coefficient ≥ 1.0 indicates high risk.

[0146] After determining the risk level of each element, continuous risk regions are identified for medium-risk and high-risk finite element elements. Based on the topological adjacency relationship between finite element mesh elements, any risk element that has not yet been classified into regions is used as the starting element. Adjacent risk elements with common surfaces are continuously searched, and interconnected risk elements are divided into the same continuous risk region.

[0147] For each continuous risk area, its spatial location, range, maximum principal tensile stress, and maximum crack risk coefficient are statistically analyzed. The maximum crack risk coefficient in the area is used as the basis for determining the risk level of the continuous risk area, thereby realizing the automatic location, range statistics, and risk classification of crack risk areas in concrete structures, providing a basis for the generation of subsequent temperature-controlled curing and crack-resistant control strategies.

[0148] Generating crack-resistant strategies is a crucial decision-making step based on crack risk prediction results, involving the synergistic optimization of temperature-controlled curing measures and structural crack-resistant measures. Its core lies in transforming risk indicators obtained from the coupled analysis of temperature field, stress field, and shrinkage / creep into specific, actionable curing and crack-resistant control instructions, thereby forming a comprehensive crack prevention and control plan for different regions, ages, and risk levels. The crack-resistant strategy mainly includes two parts: a temperature-controlled curing strategy and a crack-resistant control strategy. By reducing the peak internal temperature of concrete, minimizing the temperature difference between the inside and surface, inhibiting shrinkage deformation, and weakening constraint stress, it achieves proactive control of concrete crack risk.

[0149] Temperature-controlled curing strategies are primarily used to regulate the temperature development process during concrete construction and curing. The goal is to reduce the temperature gradient and cooling rate, thereby minimizing temperature stress caused by hydration heat release. Temperature-controlled curing parameters include the initial placement temperature, insulation layer thickness, cooling intensity, demolding time, spraying frequency, and curing duration. Specifically, initial placement temperature control is determined by analyzing the development patterns of ambient temperature and hydration heat release. When the control system predicts that the internal temperature peak exceeds a preset threshold, it automatically generates a control scheme to reduce the initial placement temperature, such as using aggregate pre-cooling, cooling mixing water, or nighttime pouring. For areas with rapid internal temperature increases, the control system automatically increases the insulation layer thickness and extends the formwork covering time to slow the rate of temperature difference development between the inside and outside.

[0150] During the cooling phase, the central control unit analyzes the rate of temperature change and temperature gradient distribution in real time based on a digital twin finite element model. When it predicts that a local area may experience significant tensile stress due to rapid cooling, it automatically reduces the cooling water circulation rate or slows down the removal of the insulation layer, allowing the heat inside the concrete structure to be released gradually and preventing drastic temperature shrinkage. Simultaneously, the temperature-controlled curing system dynamically adjusts the intensity of spray curing based on real-time ambient temperature, wind speed, and solar radiation conditions. For example, when the outside temperature rises or wind speed increases, leading to a higher surface evaporation rate, the system automatically increases the spraying frequency and shortens the spraying interval; when the ambient temperature drops rapidly at night, it automatically strengthens the insulation covering measures to reduce the surface cooling rate.

[0151] The temperature-controlled curing strategy is generated based on temperature field simulation results and a temperature control rule base. The rule base is established according to engineering specifications, material properties, and historical engineering data, and includes rules for maximum temperature control, surface-to-internal temperature difference control, and cooling rate control. The temperature control rule base is established using conditional rules; each control rule includes a rule number, input indicators, trigger conditions, execution actions, control parameters, rule priority, and release conditions. Before concrete pouring, the rule base is initialized based on engineering design documents, concrete material performance test results, construction temperature control targets, and construction organization plans, and then stored in the strategy database of the central processing unit.

[0152] The input indicators for the temperature control rule base include one or more of the following: the highest internal temperature of the concrete, the temperature difference between the inside and the surface, the cooling rate, the ambient temperature, the relative humidity, and the wind speed. The central processing unit reads the temperature field prediction results from the digital twin finite element model and the real-time monitoring data from the site at preset time intervals, and compares each input indicator with the trigger conditions in the rule base item by item.

[0153] For example, rule R1 is used for maximum temperature control. When the predicted maximum internal temperature of the concrete reaches the maximum temperature control value set by the project, a cooling control action is triggered, outputting the start status of the cooling system, the flow rate of the cooling medium, or the adjustment command for the temperature of the concrete entering the formwork. Rule R2 is used for internal and external surface temperature difference control. When the predicted internal and external surface temperature difference reaches the internal and external surface temperature difference control value set by the project, a control action is triggered to strengthen the insulation and extend the insulation time. Rule R3 is used for cooling rate control. When the predicted cooling rate reaches the upper limit of the allowable cooling rate, the operating intensity of the cooling system is reduced or the speed of removing the insulation layer is slowed down. Rule R4 is used for humidity control. When the relative humidity of the concrete surface is lower than the target humidity range, the spraying frequency is increased or the duration of a single spray is extended.

[0154] Each rule has a corresponding release condition. Once the corresponding control index returns to the safe range and continues to meet the release condition, the central processing unit terminates the corresponding enhanced control measures. This dynamically generates a temperature-controlled maintenance strategy based on temperature field prediction results and real-time environmental parameters.

[0155] The crack control strategy primarily targets areas of structural stress concentration and crack-sensitive regions for focused protection. Its goal is to reduce constraint stress and shrinkage stress, thereby slowing crack propagation. Crack control measures include locally enhanced insulation and moisture retention, extended wet curing time, optimized formwork removal sequence, adjusted construction sequence, and focused protection of high-risk areas. Specifically, when the digital twin finite element model identifies a crack risk coefficient for a corresponding finite element element at a medium or high risk level, the central control unit automatically increases the insulation and moisture retention level for that area and increases the intensity of localized spraying. For areas with a crack risk coefficient reaching a high risk level, measures such as delaying formwork removal time, increasing the number of covering layers, or setting up localized buffer insulation layers can reduce temperature stress concentration.

[0156] To address the long-term cracking risk caused by shrinkage and creep, the central control unit dynamically adjusts the curing plan based on the development patterns of shrinkage strain and constraint stress. When the constraint tensile stress caused by shrinkage deformation causes the cracking risk in the corresponding area to reach a medium-risk level, the duration of wet curing is extended and the insulation and moisture retention intensity is increased; when the cracking risk reaches a high-risk level, further enhanced protective measures such as delayed formwork removal and strengthened local insulation and moisture retention are adopted.

[0157] During the generation of crack control strategies, the prevention and control system establishes a decision-making mechanism based on a rule base. This rule base is pre-established based on concrete material properties, crack control requirements, and engineering experience. Each rule includes input indicators, triggering conditions, risk levels, and corresponding control measures. The prevention and control system matches indicators such as crack risk coefficients, shrinkage strain, and risk area range obtained from the digital twin finite element model with the triggering conditions in the rule base, and generates corresponding crack control measures based on the matching results. When the crack risk reaches a medium-risk level, the insulation and moisture retention intensity and monitoring frequency of the corresponding area are increased; when the crack risk reaches a high-risk level, enhanced protective measures such as localized key temperature control, extended wet curing time, and delayed formwork removal are initiated. When multiple rules are simultaneously met, the control rule with the higher risk level is prioritized.

[0158] The generated temperature control curing strategy and crack resistance control strategy are finally output in the form of structured data, including parameters such as target temperature range, allowable cooling rate, target humidity range, spraying cycle, heat preservation coverage time, demolding sequence and key protection areas, and are sent to the intelligent curing control system through the industrial Internet of Things platform.

[0159] The intelligent maintenance and control system automatically coordinates the operation of the cooling system, spraying system, and insulation covering system according to strategy instructions, forming a closed-loop crack prevention and control system that integrates real-time monitoring, risk analysis, strategy generation, equipment execution, and dynamic feedback. This enables intelligent, precise, and proactive control of crack risks in large-volume concrete.

[0160] During the water storage phase, thin-walled, large-volume concrete pools are subjected to hydrostatic pressure, which creates a new stress state between the water-facing and back surfaces of the pool wall. In some areas, the combined effects of residual temperature stress, water pressure, and shrinkage and creep may lead to new areas of tensile stress concentration, thereby inducing through cracks or leakage cracks.

[0161] In some feasible examples, the prevention and control methods not only predict and control temperature cracks during the concrete construction and curing stages, but also further cover the water storage test, crack repair and re-inspection stages after demolding and curing. By establishing a digital twin closed-loop control system covering the entire process of water storage, leakage, crack identification, repair and re-inspection, continuous monitoring and dynamic repair of crack risks in the early service stage of thin-walled large-volume concrete water tanks can be achieved.

[0162] Specifically, it includes:

[0163] S6. Based on the finite element model, a fluid-structure interaction analysis model is further established. The water storage load is applied in stages to the pool wall and bottom plate structure as an external force to simulate the development process of structural stress and deformation under water pressure. Temperature data, strain data and environmental parameter data of the concrete structure are automatically collected to dynamically update and calibrate the parameters of the finite element model. The collected data on pool wall displacement, strain and crack changes are simultaneously fed back to the digital twin platform in real time.

[0164] S7. During the water storage monitoring process, based on the comparison and analysis of real-time collected data and finite element prediction results, high-risk areas in the water storage stage are automatically identified and crack early warning information is generated; the crack propagation risk can be predicted by combining the water pressure change trend and crack propagation direction.

[0165] S8. Automatically generate corresponding repair plans based on crack type, crack width, and leakage level; automatically generate grouting path, grouting pressure, and grouting duration according to crack location, and execute repair operations through intelligent grouting equipment;

[0166] S9. After the crack repair is completed, a re-inspection and secondary water storage verification stage will be carried out.

[0167] Specific implementation process:

[0168] After the concrete structure is demolded and cured, the control system enters the water storage analysis phase. Based on the finite element model of the concrete structure, the control system further establishes a fluid-structure interaction analysis model, applying the water storage load as an external force to the pool walls and bottom slab structure to simulate the development process of stress and deformation in the concrete structure under water pressure. The water storage process is carried out in stages according to actual engineering conditions, such as injecting water in stages according to a set water level. At each stage, data on pool wall displacement, strain, and crack changes are collected synchronously and fed back to the digital twin platform in real time.

[0169] During water storage monitoring, piezometers, displacement gauges, and crack monitoring sensors can be deployed in sensitive areas such as the water-facing side of the pool wall, construction joints, post-cast strips, and structural corners. Combined with visual recognition equipment, real-time changes in crack width and leakage area are acquired. The control system compares and analyzes the real-time acquired data with finite element prediction results. When the detected local crack width, leakage rate, or tensile stress exceeds a preset threshold, the control system automatically identifies it as a high-risk area during the water storage phase and generates a crack warning. Simultaneously, the control system can predict crack expansion risk by combining water pressure change trends and crack propagation direction, determining whether there is a risk of further crack expansion or penetration.

[0170] During the crack repair phase, the control system automatically generates corresponding repair plans based on crack type, crack width, and leakage level. For micro-cracks on the surface, surface sealing and polymer coating repair measures can be used; for through cracks or leaking cracks, pressure grouting is used for repair, such as injecting epoxy resin grout, cement-based penetrating crystallizing materials, or polyurethane sealing materials into the crack to restore the structural integrity and impermeability. The digital twin platform automatically generates parameters such as grouting path, grouting pressure, and grouting duration based on the crack location, and executes the repair operation through intelligent grouting equipment. During the repair process, the control system monitors the grouting pressure, grout flow rate, and crack closure in real time. When the grout diffusion range reaches the preset requirements, the grouting process is automatically terminated, thereby improving the crack repair quality and construction efficiency.

[0171] After repairs are completed, the control system proceeds to the re-inspection and secondary water storage verification phase. Specifically, a high-precision digital twin model is re-established in the repaired crack area, and a phased water injection test is conducted again to verify the stress state and impermeability of the repaired structure. During the re-inspection, the control system monitors the crack width, leakage, displacement changes, and stress development in the repaired area in real time, and compares and analyzes the data with the data before repair. If no new crack expansion is detected during the re-inspection, the leakage is lower than the preset standard, and the structural stress is within a safe range, the control system determines that the repair effect meets the requirements; if abnormal leakage or local tensile stress exceeds the limit, the system automatically regenerates the repair and reinforcement strategy and enters the next repair-re-inspection cycle.

[0172] Furthermore, the control system continuously records the temperature field, stress field, crack location, crack width, leakage, and repair parameters of the concrete structure throughout the entire process of water storage, repair, and re-inspection, forming a database of crack evolution and repair records. By comparing and analyzing the crack width, leakage, structural displacement, and stress changes before and after repair, the crack repair effect is evaluated. When the re-inspection results do not meet the preset requirements, the corresponding repair and reinforcement strategies are regenerated, and the next round of repair and re-inspection begins, thereby achieving coordinated crack control throughout the entire process from the construction stage to the water storage stage of the thin-walled, large-volume concrete water tank.

[0173] This invention also provides a digital twin crack prevention and control system for thin-walled, large-volume concrete water tanks, used to implement the aforementioned prevention and control method, comprising:

[0174] The model building module is used to build digital twin finite element models of thin-walled, large-volume concrete structures.

[0175] The data acquisition module includes temperature sensors and strain gauges embedded in the concrete structure, as well as temperature and humidity monitoring devices and environmental acquisition devices deployed on the concrete surface.

[0176] The model dynamic calibration module compares, analyzes, and corrects the real-time monitoring data acquired by the data acquisition module with the predicted values ​​of the finite element model.

[0177] The multi-field coupling analysis module, based on the finite element model, embeds temperature field analysis subroutines, shrinkage analysis subroutines, and creep analysis subroutines through the secondary development interface of finite element software.

[0178] The crack risk assessment module automatically identifies and dynamically assesses high-risk crack areas based on stress analysis results and concrete strength development patterns.

[0179] The crack resistance strategy generation module generates temperature control maintenance strategies and crack resistance control strategies based on the prediction results of the crack risk assessment module.

[0180] The intelligent temperature control curing execution module includes a coordinated cooling system, a spraying system, and an insulation covering system. Based on the strategy generated by the crack-resistant strategy generation module, it automatically coordinates and executes insulation, spraying, cooling, and formwork removal sequence measures to achieve proactive control of concrete structure temperature and cracks throughout the entire process.

[0181] The central processing unit is used to receive and process real-time monitoring data transmitted by the data acquisition module, coordinate the operation of each module, execute data assimilation algorithms and parameter optimization algorithms, generate control commands and distribute them to each execution module.

[0182] As a further optimization, the prevention and control system also includes a water storage test analysis module, which specifically includes piezometers, displacement gauges, crack monitoring sensors, and visual recognition devices deployed in sensitive areas such as the water-facing side of the pool wall, construction joints, post-pouring strips, and structural corners. The water storage test analysis module obtains the changes in crack width and leakage area in real time, predicts the risk of crack expansion, and repairs the cracks using intelligent grouting equipment.

[0183] The prevention and control method and system provided by the present invention simulates the development process of concrete hydration heat release, temperature stress and shrinkage and creep stress by constructing a finite element model that considers the coupling effect of temperature field, stress field and shrinkage and creep, and combines the monitoring data of construction stage and water injection stage to dynamically predict the crack risk of the structure throughout the entire process.

[0184] Compared with traditional crack control technologies, this invention has the following significant advantages:

[0185] 1. Closed-loop management throughout the entire process, covering the entire construction cycle.

[0186] It has achieved closed-loop management and control of the entire process from construction preparation to final acceptance. Around the key processes of concrete structure forming, a digital twin dynamic management and control system covering "pouring, curing, formwork removal, water storage test, defect repair and re-inspection" has been established, integrating the previously fragmented construction and defect handling links into an organic whole, realizing real-time monitoring and dynamic zeroing of crack risks during the construction phase.

[0187] 2. Multiphysics coupling analysis, high prediction accuracy

[0188] In thin-walled, large-volume concrete, the coupled effects of temperature, stress, and shrinkage / creep are comprehensively considered, rather than just temperature and stress. By finite element simulation of hydration heat release, temperature conduction, drying shrinkage, and creep effects, the internal stress evolution of concrete structures can be more realistically reflected, significantly improving the accuracy of identifying high-risk crack areas.

[0189] 3. Digital twin dynamic calibration to achieve virtual-real synchronization.

[0190] By utilizing pre-embedded temperature sensors, strain gauges, and environmental monitoring equipment, data is collected in real time, and model parameters are automatically corrected through data assimilation algorithms, enabling the digital model to evolve synchronously with the actual structural state. This closed-loop calibration mechanism overcomes the limitations of traditional open-loop, offline simulations, ensuring the real-time nature and reliability of the prediction results.

[0191] 4. Intelligent identification and graded early warning of crack risk

[0192] Based on the dynamic changes of stress in the digital model, the system automatically extracts units that exceed the safety threshold and identifies continuous high-stress areas by combining regional clustering algorithms. This enables the automatic location, range statistics, and risk level classification (low / medium / high) of high-risk crack areas, providing a quantitative basis for differentiated maintenance.

[0193] 5. Temperature control and crack resistance strategies are optimized in synergy, with active regulation.

[0194] It not only generates routine temperature control and curing parameters, but also automatically takes anti-cracking measures such as localized enhanced heat preservation and moisture retention, delayed formwork removal, and optimized construction joint timing for stress concentration areas, forming an active closed-loop control of "monitoring-analysis-strategy-execution-feedback" rather than passive repair after the fact.

[0195] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for preventing and controlling cracks in thin-walled, large-volume concrete water tanks using digital twin technology, characterized in that... include: S1. By integrating concrete mix proportion parameters, concrete structure geometric parameters, and environmental parameters, a digital twin finite element model of a thin-walled, large-volume concrete water tank is constructed, which considers the coupling effects of temperature field, stress field, and shrinkage and creep. The finite element model is used to simulate the hydration heat release and temperature conduction process of the concrete structure, as well as the development process of temperature stress and shrinkage and creep stress. S2. During the construction phase, temperature data, strain data, and environmental parameter data of the concrete structure are automatically collected, and the finite element model is dynamically updated and its parameters are calibrated. S3. Dynamic prediction of temperature field, stress field and crack risk of concrete structure through digital twin through finite element model; S4. Based on the dynamic prediction results of crack risk, combined with the analysis results of safety factor, peak temperature, temperature difference between inside and outside and shrinkage stress, generate crack resistance strategies including temperature control maintenance strategy and crack resistance control strategy; S5. The prevention and control system automatically coordinates crack prevention and control measures according to strategy instructions.

2. The method for preventing and controlling cracks in a thin-walled, large-volume concrete water tank using digital twin technology according to claim 1, characterized in that: The construction of the digital twin finite element model in S1 includes: S11. Obtain concrete mix proportion parameters, concrete structure geometric parameters, and environmental parameters; S12. Based on the geometric parameters of the concrete structure, a three-dimensional finite element model is established and meshed using finite element analysis software; S13. Obtain time-varying parameters of the thermal and mechanical properties of materials through experiments, and import the material property parameters into the simulation environment through a software interface; S14. Embed subroutines for temperature field analysis, shrinkage analysis, and creep analysis through the secondary development interface of finite element software to perform temperature field simulation and stress field simulation; S15. Based on environmental parameters, generate thermal boundary conditions for the finite element model through a subroutine; S16. Set the initial conditions and structural constraints of the model, configure the coupled calculation logic of temperature field, stress field and shrinkage and creep, and complete the construction of the coupled finite element model.

3. The method for preventing and controlling cracks in a thin-walled, large-volume concrete water tank using digital twin technology according to claim 2, characterized in that: In S11, concrete mix proportion parameters include cement type, water-cement ratio, aggregate gradation and admixture dosage. The geometric parameters of concrete structures include the three-dimensional dimensions, spatial shape, and boundary constraints of the components. Environmental parameters include ambient temperature, relative humidity, and wind speed.

4. The method for preventing and controlling digital twin cracks in a thin-walled, large-volume concrete water tank according to claim 3, characterized in that: In S14, the temperature field simulation adopts the three-dimensional unsteady heat conduction theory. During the analysis, the concrete hydration heat release model is written using the ABAQUS user subroutine UMATHT, and the heat release law of hydration heat release with age is loaded into the finite element calculation process. In S15, the ambient temperature and convective heat transfer coefficient are defined using the FILM user subroutine, and the thermal boundary conditions are generated.

5. The method for preventing and controlling digital twin cracks in a thin-walled, large-volume concrete water tank according to claim 4, characterized in that: When configuring the coupled calculation logic for temperature field, stress field, and shrinkage / creep in S16, the material constitutive model adopts an elastic model that considers the effects of age. The coupled calculation logic includes: Temperature field simulation: Dynamic input of thermal boundary conditions is achieved through the ABAQUS secondary development interface; and the transient thermal analysis module is used to perform numerical calculations on the entire process of heat release during concrete hydration. Stress field simulation: The temperature strain, shrinkage strain and creep strain are uniformly introduced into the finite element analysis process through the UEXPAN subroutine, and the elastic modulus is dynamically updated with age by combining the UFIELD subroutine, thus establishing a coupled analysis model of the entire process of temperature, shrinkage and creep. The stress field simulation is achieved by using a sequential coupling analysis method based on the temperature field calculation results. The temperature field results obtained at each time step are imported into the structural analysis module as load conditions, and the coupled analysis model is used to calculate the stress and deformation development process of the concrete structure under temperature action.

6. The method for preventing and controlling digital twin cracks in a thin-walled, large-volume concrete water tank according to claim 5, characterized in that: S2 includes dynamic updating and parameter calibration of the finite element model, including: S21. Send the collected temperature and strain data to the central processing unit; S22. The central processing unit adopts the weighted least squares data assimilation method to compare the real-time temperature and strain monitoring data with the finite element model prediction data at the corresponding locations, establishes the model calibration objective function, and calculates the relative error of temperature prediction between the finite element model predicted temperature and the actual measured temperature, as well as the relative error of strain prediction between the finite element model predicted strain and the actual measured strain. S23. When the relative error of the temperature prediction is greater than 2% or the relative error of the strain prediction is greater than 5%, the parameter optimization algorithm is used to iteratively correct the hydration exothermic model parameters, thermal boundary condition parameters and material performance parameters; S24. Recalculate the temperature and stress fields for the current time step using the corrected model parameters, and recalculate the relative error of temperature prediction and the relative error of strain prediction. When the relative error of temperature prediction is no greater than 2% and the relative error of strain prediction is no greater than 5%, the current finite element model is determined to meet the preset calibration requirements. The corrected model parameters are saved and used as the model parameters for subsequent time steps to achieve synchronous updates between the digital twin finite element model and the actual concrete structure state. If any prediction relative error still exceeds the corresponding preset calibration allowable error, parameter iterative correction continues until the preset calibration allowable error is met or the preset maximum number of iterations is reached.

7. The method for preventing and controlling cracks in a thin-walled, large-volume concrete water tank using digital twin technology according to claim 6, characterized in that: The temperature strain is expressed as: ； Where α is the coefficient of linear expansion of concrete, and ΔT is the temperature change; Contraction strain is expressed as: ； Where t represents the concrete age considered in the calculation; t s The concrete age at the onset of shrinkage; The age at the onset of contraction is t. s The calculation considers the shrinkage strain at an age of t; β is the nominal contraction coefficient; β is the coefficient of contraction over time. Creep strain is expressed as: ； in, For creep response, σ is the creep coefficient, σ is the current stress, and E(t) is the current elastic modulus.

8. The method for preventing and controlling digital twin cracks in a thin-walled, large-volume concrete water tank according to claim 6, characterized in that: The full-process digital twin dynamic prediction of crack risk in S3 includes: S31. Obtain the tensile strength of concrete and establish a tensile strength growth model based on the test results at different ages; S32. Set a corresponding safety factor according to the importance of the concrete structure, and use the ratio of tensile strength to safety factor as the allowable tensile stress at age t, and use this as the threshold for judging crack risk. S33. Based on the stress field results calculated using the finite element model, extract the maximum principal tensile stress σ of the i-th finite element. 1,i The ratio of the maximum principal tensile stress to the allowable tensile stress is used as the crack risk coefficient for the i-th finite element: S34. Based on the crack risk coefficient, risk identification is performed on finite element elements, and based on the topological adjacency relationship between finite element mesh elements, connectivity retrieval is performed on adjacent risk elements; adjacent risk elements with common surfaces are divided into the same continuous risk region, and the spatial location, range, maximum principal tensile stress and maximum crack risk coefficient of each continuous risk region are statistically analyzed. S35. Crack risk levels are classified according to crack risk coefficients. When the coefficient is <0.7, it is classified as low risk; when 0.7≤coefficient<1.0, it is classified as medium risk; when the coefficient is ≥1.0, it is classified as high risk. For a continuous risk area composed of multiple finite element elements, the largest crack risk coefficient in the area is used as the basis for determining the risk level of the area.

9. A method for preventing and controlling digital twin cracks in a thin-walled, large-volume concrete water tank according to any one of claims 1-8, characterized in that: It also includes water retention tests after concrete structure demolding and curing, crack repair and re-inspection, specifically including: S6. Based on the finite element model, a fluid-structure interaction analysis model is further established. The water storage load is applied in stages to the pool wall and bottom plate structure as an external force to simulate the development process of structural stress and deformation under water pressure. Temperature data, strain data and environmental parameter data of the concrete structure are automatically collected to dynamically update and calibrate the parameters of the finite element model. The collected data on pool wall displacement, strain and crack changes are simultaneously fed back to the digital twin platform in real time. S7. During the water storage monitoring process, based on the comparison and analysis of real-time collected data and finite element prediction results, high-risk areas in the water storage stage are automatically identified and crack early warning information is generated; the crack propagation risk can be predicted by combining the water pressure change trend and crack propagation direction. S8. Automatically generate corresponding repair plans based on crack type, crack width, and leakage level; automatically generate grouting path, grouting pressure, and grouting duration according to crack location, and execute repair operations through intelligent grouting equipment; S9. After the crack repair is completed, a re-inspection and secondary water storage verification stage will be carried out.

10. A digital twin crack prevention and control system for thin-walled, large-volume concrete water tanks, used to implement the prevention and control method according to any one of claims 1-9, characterized in that, include: The model building module is used to build digital twin finite element models of thin-walled, large-volume concrete structures. The data acquisition module includes temperature sensors and strain gauges embedded in the concrete structure, as well as temperature and humidity monitoring devices and environmental acquisition devices deployed on the concrete surface. The model dynamic calibration module compares, analyzes, and corrects the real-time monitoring data acquired by the data acquisition module with the predicted values ​​of the finite element model. A multi-field coupling analysis module, wherein the multi-field coupling analysis module embeds temperature field analysis subroutines, shrinkage analysis subroutines and creep analysis subroutines through the secondary development interface of finite element software; The crack risk assessment module automatically identifies and dynamically assesses high-risk crack areas based on stress analysis results and concrete strength development patterns. The crack resistance strategy generation module generates temperature control maintenance strategies and crack resistance control strategies based on the prediction results of the crack risk assessment module. The intelligent temperature control curing execution module includes a coordinated cooling system, a spraying system, and a thermal insulation covering system. Based on the strategy generated by the anti-cracking strategy generation module, it automatically coordinates and executes thermal insulation, spraying, cooling, and demolding sequence. The central processing unit is used to receive and process real-time monitoring data transmitted by the data acquisition module, coordinate the operation of each module, execute data assimilation algorithms and parameter optimization algorithms, generate control commands and distribute them to each execution module.

Images