Judgment and inversion method for internal temperature cracks of large-volume thin-wall concrete in construction and maintenance period

Abstract

The invention provides a method for judging and inverting internal temperature cracks of large-volume thin-wall concrete in the construction and maintenance period, and belongs to the technical field of constructional engineering.The method comprises the steps that a scale model and a simulation analysis method are adopted, and a scale physical model and a simulation model of large-volume bottom layer concrete are established preferentially; then, an upper-layer concrete thin-wall structure reduced scale physical model and a simulation model are established on the upper surface of the upper-layer concrete thin-wall structure model, and the whole hydration heat process of the upper-layer concrete thin-wall structure model is simulated; by comparing stress and temperature distribution characteristics of the reduced-scale physical model and the simulation model for the upper-layer concrete thin-wall structure model, the occurrence and trend of the mass concrete internal temperature crack are judged by taking temperature distribution and stress distribution as dual standards, and inversion of the internal temperature crack is realized. By comparing the temperature distribution and stress distribution of the reduced scale physical model and the simulation model, the generation characteristics of the temperature cracks in the mass concrete in the construction and maintenance period are obtained, and a design basis is provided for temperature control and crack prevention of the mass concrete in the construction and maintenance period.

Description

Technical Field

[0001] The present invention belongs to the field of construction engineering technology, in particular to a technology for temperature cracking inside large-volume concrete during the construction and curing period, and more particularly to a method for determining and inverting temperature cracks inside large-volume thin-walled concrete during the construction and curing period. Background Art

[0002] Mass concrete, a key material in modern construction, is widely used in projects such as bridges, dams, and high-rise building foundations. However, during concrete construction and curing, the cement hydration reaction releases a large amount of heat, causing the concrete's internal temperature to rise sharply, while the external temperature cools rapidly due to ambient temperature or heat dissipation, resulting in a significant internal and external temperature differential. Once the temperature stress generated by this temperature differential exceeds the concrete's tensile strength, it can easily cause internal thermal cracks. These cracks not only affect the durability and load-bearing capacity of the structure, but can also lead to secondary hazards such as leakage and steel corrosion due to crack expansion, posing a serious threat to project safety.

[0003] Traditional methods for preventing and controlling temperature cracks in large-volume concrete rely on empirical temperature control measures or prediction methods based on single numerical simulations. However, existing technologies have several limitations. Conventional numerical simulation models often use simplified assumptions, making it difficult to accurately simulate the dynamic coupling of concrete hydration heat release and temperature and stress fields during the curing period, resulting in significant deviations between predicted results and actual working conditions. In actual projects, crack risk is often assessed indirectly through temperature monitoring data, lacking a comprehensive analysis of the synergistic effects of temperature and stress fields. Existing criteria also make it difficult to accurately determine the location and propagation direction of cracks. Existing methods are incapable of tracing the causes of existing cracks, making it difficult to infer the temperature and stress distribution history from crack morphology, limiting the targeted optimization of crack prevention measures. Furthermore, existing technologies lack a systematic analysis method for the interaction between large-volume concrete and superimposed thin-walled structures during the curing period, resulting in blind spots in the assessment of structural integrity. Therefore, a method that integrates physical testing and numerical simulation, considers both temperature and stress criteria, and possesses crack inversion capabilities is urgently needed to provide a scientific basis for precise temperature control and crack prevention design in large-volume concrete.

[0004] In response to the above problems, the present invention aims to propose a method for determining and inverting internal temperature cracks in large-volume thin-walled concrete during the construction and curing period. By using a scaled model and simulation analysis method, a scaled physical model and simulation model of the large-volume bottom concrete are first established. Then, a scaled physical model and simulation model of the upper concrete thin-walled structure are established on its upper surface, and the entire hydration heat process of the upper concrete thin-walled structure model is simulated. By comparing the stress and temperature distribution characteristics of the upper concrete thin-walled structure model with the scaled physical model and the simulation model, the temperature distribution and stress distribution are used as dual standards to determine the occurrence and direction of temperature cracks in the large-volume concrete, thereby realizing the inversion of internal temperature cracks. By comparing the temperature distribution and stress distribution of the scaled physical model and the simulation model, the present invention obtains the occurrence characteristics of temperature cracks in the large-volume concrete during the construction and curing period, providing a design basis for temperature control and crack prevention of large-volume concrete during the construction and curing period.

[0005] After review, very few published patents involve BSCP pipe structure innovation and prestressed fiber winding process optimization. Some related patents are shown below:

[0006] CN113466291B discloses a method for determining and inverting temperature cracks inside large-volume concrete during the curing period. The method uses inversion technology and perspective transformation formulas to obtain temperature field information and the affected area of ​​the cracks, but is difficult to use in a "bottom-layer large-volume concrete-upper-layer thin-walled concrete" structural system. Summary of the Invention

[0007] Based on the above-mentioned problems of insufficient reliability of existing models, single monitoring and judgment criteria, and weak inversion capabilities, a method for determining and inverting internal temperature cracks in large-volume thin-walled concrete during the construction and curing period is provided, which provides a design basis for temperature control and crack prevention of large-volume concrete during the construction and curing period.

[0008] The present invention provides a method for determining and inverting temperature cracks inside large-volume thin-walled concrete during the curing period, characterized in that the method comprises the following steps:

[0009] S1: Pour the bottom mass concrete and adjust the temperature through the built-in water pipe to simulate the temperature and stress changes during the pouring process of the bottom mass concrete. The method is as follows:

[0010] S11: Tie 25mm diameter HDPE water pipes at 1 / 3 of the spacing between the main bars of the built-in steel cage in the bottom mass concrete, arranged in a serpentine pattern with a pipe spacing ≤1.5m. Run single-mode distributed temperature sensing optical fibers along the diagonal direction of the steel cage, with a spacing of 0.5m between the fibers and connected by thermal fusion. Arrange vibrating wire strain gauges in a 3×3×3m grid in the concrete core area, increasing the spacing to 1×1×1m at 50mm from the surface. After the structure initially sets, install surface strain gauges (range ±3000με, accuracy ±1%FS) and platinum resistance thermometers (range -20°C to 120°C, accuracy ±0.5°C) in a 2×2m grid.

[0011] S12: A three-dimensional solid model was constructed using ABAQUS, using the C3D8R element type and a mesh size three times the maximum aggregate particle size (typically 20 mm). Material properties included a thermal conductivity of 2.8 W / (m·K), a specific heat capacity of 900 J / (kg·K), and an ACI 209 model that varied with time in the elastic modulus. The hydration heat model used an exponential heat release function based on the Arrhenius equation, with a peak temperature set at 65 ± 3°C.

[0012] S2: Compare the key parameters of the bottom mass concrete as follows:

[0013] S21: Use the Pearson correlation coefficient method to compare the surface-internal temperature difference curve within 24 hours between the physical model and the numerical simulation. The temperature difference is allowed to deviate by ≤2°C. If the deviation exceeds the standard, adjust the heat transfer coefficient by ±10% and iterate again.

[0014] S22: Compare the main frequency components of internal temperature changes through FFT spectrum analysis, and require that the temperature change rate simulation error during the hydration exothermic stage (3-7 days after pouring) is ≤0.5℃ / h;

[0015] S23: Use digital image correlation (DIC) to measure the surface strain field and compare it with the simulation results in the whole field. The strain distribution similarity coefficient in the shrinkage stage (14-28 days after pouring) must be ≥0.85.

[0016] S24: Use the 3σ criterion to screen the temperature and stress extreme points, with the maximum temperature deviation ≤3°C and the stress deviation ≤1.5MPa;

[0017] S3: Insulate the bottom layer of bulk concrete, pour the upper layer of thin-walled concrete, and adjust the temperature through built-in water pipes. The steps are as follows:

[0018] S31: The structure is heated through water pipes buried in the massive concrete of the bottom layer, with a control accuracy of ±0.5°C and a temperature gradient of ≤15°C / m in the core area. The sampling frequency of the distributed fiber optic temperature measurement system is set to 1Hz, with a spatial resolution of 0.1m.

[0019] S32: The thickness of the upper concrete layer is controlled at 0.5-1.2m. A double-layer orthogonal distributed optical fiber is used with a longitudinal spacing of 0.3m. The surface strain gauges are arranged in a 1×1m grid and preloaded with an initial compressive strain of 50με during installation.

[0020] S33: Raise the temperature at a rate of 0.5°C / min through the bottom water pipe to maintain the temperature difference between the upper concrete surface and the ambient temperature ≤ 20°C. Use an infrared thermal imager to monitor the surface temperature uniformity in real time (temperature difference ≤ 5°C).

[0021] S34: Use a fiber Bragg grating demodulator (sampling rate 100 Hz) to record temperature and strain data and synchronously store timestamp information;

[0022] S4: Set the contact surface between the bottom layer's massive concrete and the upper layer's thin-walled concrete to simulate the temperature and stress changes during the pouring of the upper layer's thin-walled concrete. The steps are as follows:

[0023] S41: Establish a shell-solid coupling model, using S4R shell elements for the upper layer and setting 5 integration points in the thickness direction;

[0024] S42: "Hard contact" is used for normal behavior, separation is allowed, the Coulomb model is used for tangential friction, the friction coefficient is 0.8 (determined by direct shear test), and the shear stress limit is set to 0.6 times the tensile strength of concrete;

[0025] S43: The upper surface of the bottom concrete layer is selected as the main surface, and the lower surface of the upper concrete layer is specified as the secondary surface. The contact search algorithm adopts the enhanced node-surface method.

[0026] S44: Set the contact tolerance to 5% of the minimum element size (typical value 0.5mm), use the symmetric penalty function method, and take the stiffness scaling factor as 0.1;

[0027] S5: Comparison of key parameters between the physical model and numerical simulation of the upper thin-walled concrete layer is as follows:

[0028] S51: Divide the thickness into five temperature measurement layers and compare the temperature difference curves between the layers. The maximum deviation allowed is 3°C.

[0029] S52: Use the moving average method (window width 1h) to process the hydration heat release rate data, requiring the simulated curve to envelop the measured data;

[0030] S53: Comparison of stress variation characteristics of the upper thin-walled concrete surface between the physical model and the numerical simulation results to verify whether the numerical simulation can simulate the shrinkage cracking of the surface concrete;

[0031] S54: Monte Carlo method was used to perform parameter sensitivity analysis to verify the significance of temperature-stress coupling effect (p < 0.05);

[0032] S55: Set temperature measurement points every 0.1m along the height direction to verify whether the temperature transfer rate complies with Fourier's law (R 2 ≥0.95);

[0033] S56: Contact surface stress monitoring uses a piezoelectric force sensor (range 0-10 MPa, accuracy ±0.1% FS). The over-limit moment is defined as the time point when the stress first reaches 90% of the tensile strength.

[0034] S57: Set triple convergence criteria: displacement change rate <0.5%, residual force <1%, energy error <0.1%;

[0035] S6: Set up acoustic emission sensors, draw the temperature field based on the embedded optical fiber, set up high-precision cameras on the upper thin-walled concrete surface, and use the strain exceeding the concrete material cracking strain as the threshold to determine the occurrence and distribution of structural cracks. Finally, compare and invert the occurrence and distribution of structural cracks in the physical model and simulation. The steps are as follows:

[0036] S61: The acoustic emission probe uses a broadband sensor (frequency range 50kHz-1MHz), the array spacing is calculated based on a wave speed of 5000m / s, and the time difference positioning accuracy is ≤10mm;

[0037] S62: The temperature field is constructed using spatial interpolation, with a time slice interval of 1 h and an isotherm spacing of 2 °C;

[0038] S63: During the strain field fusion process, the discrete measurement points are interpolated to the finite element mesh nodes, and the residual error is controlled to ≤5με;

[0039] S64: Based on the acoustic emission detection and positioning results of cracks in the bottom mass concrete and upper thin-walled concrete, combined with the areas exceeding the cracking strain of the concrete material in the ABAQUS simulation results, the surface crack image and strain image taken by a high-precision camera set directly on the surface of the upper thin-walled concrete were combined. The three results were superimposed to finally invert the time, location and expansion path of the crack.

[0040] The step S64 further includes the following steps:

[0041] S64.1: Alignment of coordinate systems of data sources: In the acoustic emission (AE) results, elastic wave events when cracks are generated are recorded, and event coordinate point sets S in three-dimensional space are generated by multi-sensor time difference positioning (such as the time difference of arrival method). AE ={(x i ,y i ,z i ,t i ,A i)}, including position, time and amplitude information; the results of high-precision camera shooting are analyzed by DIC, and in the results, the surface displacement field is obtained by binocular or multi-camera system to generate a three-dimensional strain field point set S DIC ={(x j ,y j ,z j ,ε j ,u j )}, containing strain and displacement information; mesh node dataset S generated by finite element model (FEM) or discrete element model (DEM) FEM ={(x k ,y k ,z k ,σ k ,ε k )}, containing the stress and strain prediction values; finally, a coordinate transformation is used to map all data to a unified coordinate system;

[0042] S64.2: Multi-source data preprocessing: For acoustic emission data, wavelet transform or short-time Fourier analysis is used to separate noise frequency bands, and anomalous events are eliminated by combining amplitude thresholds and cluster analysis. For DIC data, local weighted regression or Gaussian filtering is used to eliminate speckle noise, and the strain gradient threshold is used to identify the true crack area. For simulation analysis results, the simulation results can be optimized by inverting material parameters (such as elastic modulus and fracture energy) based on the DIC measured displacement field to reduce model errors.

[0043] S64.3: Data fusion: AE event points, DIC strain field and simulation grid data are mapped to the same dense grid through Kriging interpolation (Kriging) or radial basis function (RBF) to form a three-dimensional fused dataset S fusion ={(x,y,z,A,ε,σ)}; Identify inconsistencies in multiphysics data (e.g., areas where AE events clearly contradict the simulated stress field) based on Mahalanobis distance or principal component analysis (PCA).

[0044] S64.4: Crack direction identification: Use the AE event clustering direction to determine the macroscopic crack direction through principal component analysis or tensor decomposition; combine the direction of the DIC strain concentration area and the simulated stress field gradient to verify the crack propagation path.

[0045] S64.5: Crack Width Estimation: Calculate surface crack width using DIC local displacement jumps; invert internal crack width based on the relationship between AE energy release rate G and crack width; output crack width w using node separation or phase field model in numerical simulations FEM , compared with the measured value and iteratively corrected.

[0046] Compared with the prior art, the advantages and positive effects of the present invention are:

[0047] 1. Improve the accuracy of crack determination. By comparing the physical model with the numerical simulation results in multiple key parameters (such as temperature difference, temperature change characteristics, stress change characteristics, maximum temperature and stress difference, temperature rise characteristics along the height, contact surface stress change and exceedance time, stress exceedance time, etc.), the accuracy of the numerical simulation can be fully and carefully verified, thus providing a more reliable basis for crack determination. Utilizing a variety of monitoring methods such as acoustic emission sensors, distributed temperature sensing fibers, strain gauges, and high-precision cameras, temperature, strain, cracks, and other information inside and on the surface of concrete can be obtained from different angles. This information is then comprehensively analyzed and inverted, greatly improving the accuracy of crack occurrence and distribution determination and avoiding the errors and limitations that may exist in a single monitoring method.

[0048] 2. Improve the reliability of crack inversion. The data generated by acoustic emission, DIC analysis, and finite element or discrete element models are mapped to a unified coordinate system through coordinate transformation, ensuring the spatial consistency of multi-source data, providing an accurate spatial basis for subsequent data fusion and crack inversion, and avoiding errors caused by inconsistent coordinate systems. Wavelet transform or short-time Fourier analysis is performed on acoustic emission data to separate noise frequency bands and eliminate abnormal events; local weighted regression or Gaussian filtering is used on DIC data to eliminate speckle noise and identify real crack areas; simulation analysis results are optimized by inverting material parameters based on the DIC measured displacement field. These preprocessing measures effectively improve the quality and reliability of the data and reduce the interference of noise and abnormal data on crack inversion. Spatial interpolation and matching methods are used to map data from different sources onto the same dense grid to form a three-dimensional fused data set, and inconsistent points are identified through Mahalanobis distance or principal component analysis, further improving the accuracy and reliability of data fusion and providing more accurate data support for crack direction identification and width estimation. The macroscopic direction of the cracks is determined by combining the clustering direction of AE events with principal component analysis or tensor decomposition, and the crack propagation path is verified by combining the direction of the DIC strain concentration area and the simulated stress field gradient. At the same time, the crack width is estimated through a variety of methods (such as calculating the surface crack width based on the DIC local displacement jump, inverting the internal crack width based on the AE energy release rate, and iteratively correcting the crack width output by numerical simulation by comparing it with the measured value). The crack direction and width are determined from multiple angles, greatly improving the reliability of the crack inversion results.

[0049] 3. Optimize temperature control during construction. During the pouring of large-volume concrete on the bottom layer, the temperature is adjusted through built-in water pipes, and ABAQUS is used to simulate the entire process of hydration heat release during curing. The accuracy of the numerical simulation is corrected by comparing the physical model and the numerical simulation results, so that the temperature changes of the bottom concrete can be controlled more accurately to prevent cracks caused by excessive temperature differences. When pouring the upper thin-walled concrete, the bottom layer is heated through the built-in water pipes in the bottom concrete, so that the temperature is transferred to the upper thin-walled concrete to ensure that it does not lose temperature and crack. The temperature and strain changes inside the upper concrete are recorded through pre-buried distributed temperature sensing optical fibers and strain gauges. At the same time, the temperature changes and stress changes during the pouring of the upper thin-walled concrete are simulated, further optimizing the temperature control measures of the upper concrete and effectively reducing the risk of cracks in the upper thin-walled concrete. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0051] Figure 1 A flow chart of a method for determining and inverting internal temperature cracks in large-volume thin-walled concrete during the curing period provided in Example 1;

[0052] Figure 2 A schematic diagram of a DIC-based upper thin-wall structure surface crack location provided in Examples 1 and 2;

[0053] Figure 3 A schematic diagram of the distribution and direction of cracks on the surface of an upper thin-walled structure using multi-source monitoring data fusion provided in Examples 1 and 2;

[0054] Figure 4 A schematic diagram of a crack on the surface of an upper thin-walled structure based on acoustic emission crack location provided in Examples 1 and 2;

[0055] Figure 5 A schematic diagram of crack distribution and strike inversion on the surface of an upper thin-walled structure based on acoustic emission crack location provided in Examples 1 and 2;

[0056] Figure 6 A schematic diagram of the pinpointing of cracks on the surface of an upper thin-walled structure using multi-source monitoring data fusion provided in Examples 1 and 2;

[0057] Figure 7 A schematic diagram of the distribution and direction of cracks on the surface of an upper thin-walled structure using multi-source monitoring data fusion provided in Examples 1 and 2; DETAILED DESCRIPTION

[0058] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other in the absence of conflict.

[0059] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways than those described herein. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0060] Example 1: This example aims to implement a method for determining and inverting temperature cracks in large-volume thin-walled concrete during the curing period. During implementation, the following steps need to be followed to ensure the correctness of the implementation:

[0061] 1. Pour the bottom layer of mass concrete and regulate the temperature through internal water pipes to simulate the temperature and stress changes during the pouring process. Water pipes are tied to the internal steel cage of the bottom layer of mass concrete, and distributed temperature sensing optical fibers and strain gauges are embedded inside. After the structure has initially set, strain gauges and thermometers are arranged on the surface. Using ABAQUS, modeling is performed according to the actual size and material properties of the physical model to simulate the entire process of hydration and heat release during the curing period of the bottom layer of mass concrete.

[0062] 2. Compare key parameters of the underlying mass concrete, and compare the temperature difference between the surface and interior of the underlying mass concrete from the physical model and numerical simulation results to calibrate the accuracy of the heat conduction in the numerical simulation; compare the temperature variation characteristics of the interior of the underlying mass concrete from the physical model and numerical simulation results to calibrate the accuracy of the numerical simulation of concrete hydration heat release; compare the stress variation characteristics of the surface of the underlying mass concrete from the physical model and numerical simulation results to verify whether the numerical simulation can simulate the shrinkage cracking of the surface concrete; compare the maximum temperature and stress difference of the underlying mass concrete from the physical model and numerical simulation results to verify whether the numerical simulation is correct;

[0063] 3. Insulate the bottom layer of bulk concrete, pour the upper layer of thin-walled concrete, and regulate the temperature through built-in water pipes. Heat the structure through water pipes embedded within the bottom layer of bulk concrete, and control the internal temperature of the structure through embedded distributed temperature sensing optical fibers. Pour the upper layer of thin-walled concrete on the upper surface of the bottom layer of bulk concrete, embed distributed temperature sensing optical fibers and strain gauges within it, and wait for the structure to initially set before placing strain gauges and thermometers on the surface. During the curing period of the upper layer of thin-walled concrete, heat the bottom layer of bulk concrete through built-in water pipes, thereby transferring temperature from the bottom layer of bulk concrete to the upper layer of thin-walled concrete to prevent dehumidification and cracking. Changes in temperature and strain within the upper layer of concrete during the curing process are recorded through distributed temperature sensing optical fibers and strain gauges within the upper layer of concrete.

[0064] 4. Set the contact surface between the bottom layer of mass concrete and the upper layer of thin-walled concrete, simulate the temperature and stress changes during the pouring of the upper layer of thin-walled concrete, and build a model of the upper layer of thin-walled concrete. In the Property module, create a Contact property, select Hard Contact, then select Coulomb Friction, and enter a friction coefficient of 0.8. In the Interaction module, select Create, define the master and slave surfaces, and associate the contact properties. Set the contact algorithm and control the contact search tolerance (Contact Controls) to improve convergence.

[0065] 5. Compare the key parameters of the physical model and numerical simulation of the upper thin-walled concrete, compare the temperature difference between the surface and interior of the upper thin-walled concrete in the physical model and the numerical simulation results, to correct the accuracy of the heat conduction in the numerical simulation; compare the temperature change characteristics of the interior of the upper thin-walled concrete in the physical model and the numerical simulation results, to correct the accuracy of the numerical simulation of the hydration heat release of concrete; compare the stress change characteristics of the surface of the upper thin-walled concrete in the physical model and the numerical simulation results, to verify whether the numerical simulation can simulate the shrinkage cracking of the surface concrete; compare the maximum temperature and stress difference of the upper thin-walled concrete in the physical model and the numerical simulation results, to verify whether the numerical simulation is correct; compare the temperature rise characteristics of the upper thin-walled concrete along the height in the physical model and the numerical simulation results, to verify whether the concrete thermal conductivity in the numerical simulation results is correct; compare the stress change and exceeding limit time of the upper and lower concrete contact surfaces of the upper thin-walled concrete in the physical model and the numerical simulation results, to verify the correctness of the numerical simulation interface setting; compare the stress exceeding limit time of the upper thin-walled concrete in the physical model and the numerical simulation results, to verify the correctness of the numerical simulation;

[0066] 6. Set up acoustic emission sensors, draw the temperature field based on the pre-buried optical fiber, set up a high-precision camera on the surface of the upper thin-walled concrete, and use the threshold value of exceeding the cracking strain of the concrete material to determine the occurrence and distribution of structural cracks. Finally, compare and invert the occurrence and distribution of structural cracks in the physical model and simulation. Arrange an acoustic emission probe array on the surface of the upper thin-walled concrete, and arrange an encrypted acoustic emission probe array at the interface between the upper thin-walled concrete and the bottom large-volume concrete; draw the internal temperature field of the upper thin-walled concrete based on the pre-buried distributed temperature sensing optical fiber, describe the temperature monitoring value of each point inside the concrete based on the spatial position of the buried distributed temperature sensing optical fiber, and then draw the temperature field based on the change over time. A time-varying diagram of the internal temperature distribution of the upper thin-walled concrete layer based on fiber optic sensing data; a time-varying strain curve of the strain measuring points inside the upper thin-walled concrete layer was drawn based on the embedded strain gauges, and integrated with the strain cloud map from the ABAQUS simulation results to locate a relatively detailed internal strain field of the concrete; the detection and location results of cracks in the underlying bulk concrete and upper thin-walled concrete layer based on acoustic emission, combined with the areas exceeding the cracking strain of the concrete material shown in the ABAQUS simulation results, and the surface crack map and strain map captured by a high-precision camera set directly on the surface of the upper thin-walled concrete layer. The three results were superimposed to ultimately invert the time, location, and expansion path of the crack.

[0067] Example 2: This example aims to implement a specific method for inverting temperature cracks inside large-volume thin-walled concrete during the curing period. During the implementation process, the following steps need to be followed to ensure the correctness of the implementation:

[0068] 1. Alignment of the coordinate system of the data source: In the acoustic emission (AE) results, the elastic wave events when the cracks are generated are recorded, and the event coordinate point set S in the three-dimensional space is generated by multi-sensor time difference positioning (such as the wave arrival time difference method). AE ={(x i ,y i ,z i ,t i ,A i )}, including position, time and amplitude information; the results of high-precision camera shooting are analyzed by DIC, and in the results, the surface displacement field is obtained by binocular or multi-camera system to generate a three-dimensional strain field point set S DIC ={(x j ,y j ,z j ,ε j ,u j )}, containing strain and displacement information; mesh node dataset S generated by finite element model (FEM) or discrete element model (DEM) FEM ={(x k ,y k ,z k ,σk ,ε k )}, containing the stress and strain prediction values; finally, a coordinate transformation is used to map all data to a unified coordinate system;

[0069] 2. Multi-source data preprocessing: For acoustic emission data, wavelet transform or short-time Fourier analysis is used to separate noise frequency bands, and anomalous events are eliminated by combining amplitude thresholding and cluster analysis. For DIC data, local weighted regression or Gaussian filtering is used to eliminate speckle noise, and strain gradient thresholding is used to identify true crack areas. For simulation analysis results, the simulation results can be optimized by inverting material parameters (such as elastic modulus and fracture energy) based on the measured DIC displacement field to reduce model errors.

[0070] 3. Data fusion strategy: spatial interpolation and matching: AE event points, DIC strain field and simulation grid data are mapped to the same dense grid through Kriging interpolation (Kriging) or radial basis function (RBF) to form a three-dimensional fusion data set S fusion ={(x,y,z,A,ε,σ)}; Identify inconsistencies in multiphysics data (e.g., areas where AE events clearly contradict the simulated stress field) based on Mahalanobis distance or principal component analysis (PCA).

[0071] 4. Crack direction identification: Using the clustering direction of AE events, principal component analysis or tensor decomposition is used to determine the macroscopic crack direction. The crack propagation path is verified by combining the direction of the DIC strain concentration area and the simulated stress field gradient.

[0072] 5. Crack width estimation: Calculate surface crack width through DIC local displacement jump; invert internal crack width based on the relationship between AE energy release rate G and crack width; output crack width w through node separation or phase field model in numerical simulation FEM , compared with the measured value and iteratively corrected.

Claims

1. A method for determining and inverting temperature cracks inside large-volume thin-walled concrete during the curing period, characterized in that: The method comprises the following steps: S1: Pour the bottom mass concrete and adjust the temperature through the built-in water pipe to simulate the temperature and stress changes during the pouring process of the bottom mass concrete. The method is as follows: S11: Tie water pipes to the built-in steel cage of the massive concrete on the bottom floor, embed distributed temperature sensing optical fibers and strain gauges inside, and arrange strain gauges and temperature gauges on the surface after the structure has initially set. S12: Using ABAQUS, we modeled the entire process of hydration heat release during the curing of the bottom mass concrete according to the actual size and material properties of the physical model. S2: Compare the key parameters of the bottom mass concrete as follows: S21: Compare the temperature difference between the surface and interior of the bottom mass concrete of the physical model and the numerical simulation results to correct the accuracy of the heat conduction of the numerical simulation; S22: Comparison of the temperature variation characteristics of the bottom mass concrete between the physical model and the numerical simulation results to calibrate the accuracy of the numerical simulation of concrete hydration heat release; S23: Comparison of stress variation characteristics of the bottom mass concrete surface layer between the physical model and the numerical simulation results to verify whether the numerical simulation can simulate the shrinkage cracking of the surface concrete; S24: Compare the maximum temperature and stress difference of the bottom mass concrete between the physical model and the numerical simulation results to verify whether the numerical simulation is correct; S3: Insulate the bottom layer of bulk concrete, pour the upper layer of thin-walled concrete, and adjust the temperature through built-in water pipes. The steps are as follows: S31: The structure is heated through water pipes buried inside the massive concrete of the bottom layer, and the internal temperature of the structure is controlled through embedded distributed temperature sensing optical fibers; S32: Pour an upper layer of thin-walled concrete on the upper surface of the bottom mass concrete, embed distributed temperature sensing optical fibers and strain gauges inside the upper layer, and arrange strain gauges and temperature gauges on the surface after the structure has initially set; S33: During the curing period of the upper thin-walled concrete, the bottom mass concrete is heated through the water pipes built into the bottom mass concrete, thereby transferring the temperature from the bottom mass concrete to the upper thin-walled concrete to ensure that the upper thin-walled concrete does not lose temperature and crack; S34: The temperature and strain changes inside the upper concrete layer during the curing process are recorded using distributed temperature sensing optical fibers and strain gauges inside the upper concrete layer. S4: Set the contact surface between the bottom layer's massive concrete and the upper layer's thin-walled concrete to simulate the temperature and stress changes during the pouring of the upper layer's thin-walled concrete. The steps are as follows: S41: Establish the upper thin-walled concrete model; S42: Create a contact property in the property module, select hard contact, then select coulomb friction, and enter the friction coefficient as 0.8; S43: Select create in the interaction module, define the master surface and slave surface, and associate the contact properties; S44: Set the contact algorithm and control the contact search tolerance (Contact Controls) to improve convergence; S5: Comparison of key parameters between the physical model and numerical simulation of the upper thin-walled concrete layer is as follows: S51: Comparison of the temperature difference between the surface and interior of the upper thin-walled concrete layer between the physical model and the numerical simulation results to correct the accuracy of the heat conduction in the numerical simulation; S52: Comparison of the temperature variation characteristics of the upper thin-walled concrete between the physical model and the numerical simulation results to calibrate the accuracy of the numerical simulation of concrete hydration heat release; S53: Comparison of stress variation characteristics of the upper thin-walled concrete surface between the physical model and the numerical simulation results to verify whether the numerical simulation can simulate the shrinkage cracking of the surface concrete; S54: Compare the maximum temperature and stress difference of the upper thin-wall concrete between the physical model and the numerical simulation results to verify whether the numerical simulation is correct; S55: Comparison of the temperature rise characteristics of the upper thin-walled concrete in the physical model and the numerical simulation results to verify whether the concrete thermal conductivity in the numerical simulation results is correct; S56: Comparison of stress changes and over-limit moments at the contact surface of the upper and lower thin-walled concrete layers between the physical model and the numerical simulation results is used to verify the correctness of the numerical simulation interface settings; S57: Compare the stress exceeding moment of the upper thin-wall concrete of the physical model and the numerical simulation results to verify the correctness of the numerical simulation; S6: Set up acoustic emission sensors, draw the temperature field based on the embedded optical fiber, set up high-precision cameras on the upper thin-walled concrete surface, and use the strain exceeding the concrete material cracking strain as the threshold to determine the occurrence and distribution of structural cracks. Finally, compare and invert the occurrence and distribution of structural cracks in the physical model and simulation. The steps are as follows: S61: Arrange an acoustic emission probe array on the surface of the upper thin-walled concrete layer, and arrange a denser array of acoustic emission probes at the interface between the upper thin-walled concrete layer and the bottom mass concrete layer; S62: Based on the pre-buried distributed temperature sensing optical fiber, the temperature field inside the upper thin-walled concrete is mapped. Based on the spatial location of the buried distributed temperature sensing optical fiber, the temperature monitoring value of each point inside the concrete is described. Then, based on the change over time, a time-varying graph of the temperature distribution inside the upper thin-walled concrete based on the optical fiber sensing data is mapped. S63: Based on the embedded strain gauges, the strain time-varying curves of the strain measurement points inside the upper thin-walled concrete are drawn. This is then integrated with the strain cloud diagram from the ABAQUS simulation results to locate a more detailed strain field inside the concrete. S64: Based on the acoustic emission detection and positioning results of cracks in the bottom mass concrete and upper thin-walled concrete, combined with the areas exceeding the cracking strain of the concrete material in the ABAQUS simulation results, the surface crack image and strain image taken by a high-precision camera set directly on the surface of the upper thin-walled concrete were combined. The three results were superimposed to finally invert the time, location and expansion path of the crack.

2. The method for determining and inverting temperature cracks in large-volume thin-walled concrete during the curing period according to claim 1 is characterized in that: The step S64 further includes the following steps: S64.1: Alignment of the coordinate system of the data source: In the acoustic emission (AE) results, record the elastic wave events when the crack is generated, and generate the event coordinate point set S in three-dimensional space through multi-sensor time difference positioning (such as the wave arrival time difference method). AE ={(x i ,y i ,z i ,t i ,A i )}, including position, time and amplitude information; the results of high-precision camera shooting are analyzed by DIC, and in the results, the surface displacement field is obtained by binocular or multi-camera system to generate a three-dimensional strain field point set S DIC ={(x j ,y j ,z j ,ε j ,u j )}, containing strain and displacement information; mesh node dataset S generated by finite element model (FEM) or discrete element model (DEM) FEM ={(x k ,y k ,z k ,σ k ,ε k )}, containing the stress and strain prediction values; finally, a coordinate transformation is used to map all data to a unified coordinate system; S64.2: Multi-source data preprocessing: For acoustic emission data, use wavelet transform or short-time Fourier analysis to separate noise frequency bands, and combine amplitude thresholding and cluster analysis to eliminate abnormal events; For DIC data, local weighted regression or Gaussian filtering is used to eliminate speckle noise, and the real crack area is identified by strain gradient threshold. For simulation analysis results, the simulation results can be optimized by inverting material parameters (such as elastic modulus and fracture energy) based on the DIC measured displacement field to reduce model errors. S64.3: Data fusion strategy: spatial interpolation and matching: AE event points, DIC strain field and simulation grid data are mapped to the same dense grid through Kriging interpolation (Kriging) or radial basis function (RBF) to form a three-dimensional fusion data set S fusion ={(x,y,z,A,ε,σ)}; Identify inconsistencies in multiphysics data (e.g., areas where AE events clearly contradict the simulated stress field) based on Mahalanobis distance or principal component analysis (PCA). S64.4: Crack direction identification: Use the AE event clustering direction to determine the macroscopic crack direction through principal component analysis or tensor decomposition; combine the direction of the DIC strain concentration area and the simulated stress field gradient to verify the crack propagation path. S64.5: Crack Width Estimation: Calculate surface crack width using DIC local displacement jumps; invert internal crack width based on the relationship between AE energy release rate G and crack width; output crack width w using node separation or phase field model in numerical simulations FEM , compared with the measured value and iteratively corrected.

Images