A method for co-controlling creep prediction and crack control in UHPC for ultra-large span hybrid structures

Abstract

This invention discloses a method for predicting UHPC creep and co-controlling cracks in ultra-large span hybrid structures, relating to the field of concrete creep deformation prediction technology. The invention selects UHPC samples at various ages and collects UHPC creep data using a creep meter based on these samples. Simultaneously, it processes the collected UHPC creep data using data processing methods. After processing, it predicts creep based on the processed UHPC creep data using data analysis methods. Furthermore, it sets a measurement area for the ultra-large span hybrid structure and measures crack data in real time within this area. After measurement, it predicts crack propagation and self-healing based on the creep-predicted UHPC data. A prediction model is then constructed based on the predicted crack propagation and self-healing. Finally, based on the constructed prediction model, it co-controls cracks within the measurement area of ​​the ultra-large span hybrid structure, improving the real-time performance of crack control.

Description

Technical Field

[0001] This invention relates to the field of concrete creep deformation prediction technology, specifically a method for UHPC creep prediction and crack synergistic control applicable to ultra-large span hybrid structures. Background Technology

[0002] Creep is an inherent time-varying characteristic of concrete materials. Concrete creep refers to the phenomenon that the deformation of a concrete structure will increase over time under continuous load. Creep is also the main cause of prestress loss, cross-sectional cracking, and significant increase in mid-span deflection in long-span prestressed concrete bridges.

[0003] Existing technology, such as the invention patent application with publication number CN121031104A, discloses a method and system for predicting creep deformation of coal gangue aggregate concrete. The method includes: using the Counter model based on two-phase composite material theory, constructing an expression for the influence of coal gangue aggregate on the final creep value and creep development trend of concrete, and introducing the corresponding expression into the European standard EC2 creep model to finally establish a creep prediction model for coal gangue aggregate concrete. This invention introduces an expression for the final creep value amplification factor of concrete and incorporates this amplification factor into the European standard EC2 model. The aforementioned creep deformation model expression and amplification factor are not mentioned in existing methods, filling a gap in the prediction of creep performance of coal gangue aggregate concrete; and experiments have proven the effectiveness of the creep deformation prediction model for coal gangue aggregate concrete designed in this invention.

[0004] As can be seen from the above schemes, most current predictions of concrete creep focus on the performance of concrete and fail to consider the self-healing process of concrete, which has certain limitations. Summary of the Invention

[0005] The purpose of this invention is to provide a method for creep prediction and crack co-control of UHPC applicable to ultra-large span hybrid structures, which solves the problems existing in the background art.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a method for creep prediction and crack synergistic control of UHPC applicable to ultra-large span hybrid structures, specifically including the following steps: S1. Select UHPC samples of different ages, and collect UHPC creep data using a creep meter based on the selected UHPC samples. S2. The collected UHPC creep data is processed using data processing methods to obtain the processed UHPC creep data; S3. Based on the processed UHPC creep data, creep prediction is performed on the processed UHPC creep data using data analysis methods to obtain the UHPC data after creep prediction. S4. Set the measurement area of ​​the ultra-large span hybrid structure, and measure the crack data in real time within the set measurement area. After the measurement is completed, predict the crack propagation and self-healing based on the UHPC data after creep prediction, and construct crack propagation and self-healing models based on the predicted crack propagation and self-healing. S5. Based on the construction of crack propagation and crack self-healing models, collaborative control of cracks in the measurement area of ​​ultra-large span hybrid structures is carried out.

[0007] Preferably, the step of selecting UHPC samples of various ages and collecting UHPC creep data using a creep meter based on the selected UHPC samples includes the following steps: Set the selected UHPC samples to be of the same size, place a strain gauge at the axis of the selected UHPC sample, and set multiple measurement points around the strain gauge. Ambient temperature, ambient humidity, and initial internal stress of selected UHPC samples were collected. The initial internal stress data of selected UHPC samples under different ambient temperatures and humidity were recorded by controlling the variable method. The selected UHPC samples were numbered, and then creep experiments were performed on the selected UHPC samples after the numbering was completed. During the experiment, the selected UHPC sample was measured by the corresponding creep meter installed at the set measurement point. The creep data of UHPC was obtained by applying external stress to the selected UHPC sample through the creep meter. Three creep specimens were stacked on each creep meter, and a ball joint support was placed on the top of each stack of specimens to ensure that the UHPC specimens were subjected to axial force.

[0008] Preferably, the step of processing the collected UHPC creep data to obtain processed UHPC creep data includes the following steps: The collected UHPC creep data were time-series arranged based on the measurement time of the creep meter, and the UHPC creep data sequence was obtained after the arrangement was completed. The UHPC creep data sequence is traversed sequentially by data traversal to remove missing data in the UHPC creep data sequence; The missing data in the UHPC creep data sequence is located, the average value of the two adjacent UHPC creep data is calculated, and the missing data in the UHPC creep data sequence is filled in based on the calculated average value to obtain the processed UHPC creep data.

[0009] Preferably, the step of performing creep prediction on the processed UHPC creep data using data analysis methods to obtain the creep-predicted UHPC data includes the following steps: S31. Based on the processed UHPC creep data, the processed UHPC creep data is analyzed using data analysis methods to determine the creep coefficient; The processed UHPC creep data were fitted using the linear least squares method, and the creep coefficient of the corresponding medium inside the refrigeration unit was calculated based on the fitting results. Based on the processed UHPC creep data, the coordinates of the UHPC creep data are set as follows: ; in, This represents the first set of processed UHPC creep data and the corresponding time. Indicates the first UHPC creep data after group processing and corresponding time; The formula for the linear relationship between two adjacent sets of coordinates is defined as follows: ; The linear least squares fitting formula is shown below: ; in, This represents the fitted UHPC creep data. Represents the slope parameter. Represents linear parameters; The slope parameter is set to the creep coefficient of the processed UHPC creep data; S32. Based on the determined creep coefficient, creep prediction is performed on the processed UHPC creep data using data prediction methods to obtain the creep-predicted UHPC data.

[0010] Preferably, the step of performing creep prediction on the processed UHPC creep data based on a determined creep coefficient, to obtain the creep-predicted UHPC data, includes the following steps: S321. Determine the creep prediction model based on the determined creep coefficient; During the process of applying external stress to the selected UHPC sample by the creep meter, a creep prediction model is set up and determined based on the determined creep coefficient. The creep prediction model is set as follows: ; in, Indicates UHPC samples in The measured elastic modulus, These represent different ages of the UHPC samples. This represents the fundamental creep caused by unit stress during the UHPC sample measurement process. This represents the drying creep caused by unit stress during the UHPC sample measurement process. Indicates the error parameter. Represents the creep coefficient during the UHPC sample measurement process; S322. Based on the creep prediction model, creep prediction is performed on the processed UHPC creep data through data prediction to obtain the creep-predicted UHPC data.

[0011] Preferably, the step of performing creep prediction on the processed UHPC creep data based on the creep prediction model to obtain the creep-predicted UHPC data includes the following steps: The processed UHPC creep data and creep prediction model are input into a BP neural network. The structure of the BP neural network includes an input layer, a hidden layer, and an output layer. The prediction process of the BP neural network includes a forward propagation process and an error back-adjustment process. The creep prediction model is set as the training function of a BP neural network; set up The weights from the input layer to the hidden layer; The forward propagation process of the BP neural network is defined as follows: the processed UHPC creep data is input into the hidden layer of the BP neural network through the input layer of the BP neural network. The hidden layer of the BP neural network processes the input processed UHPC creep data based on the training function of the BP neural network and the weights from the input layer to the hidden layer. After the processing is completed, the processing result is transmitted to the output layer of the BP neural network to obtain the UHPC data after creep prediction. Set an error threshold between the output layer and the expected value. When the error between the output layer and the expected value is greater than or equal to the set error threshold, adjust the weights from the hidden layer to the output layer through the error back-adjustment process and perform iterative calculations until the error converges. Output the corresponding BP neural network model and the UHPC data after creep prediction at the next time step.

[0012] Preferably, the step of setting a measurement area for the ultra-large span hybrid structure, measuring crack data in real time within the set measurement area, predicting crack propagation and self-healing based on the UHPC data after creep prediction, and constructing crack propagation and self-healing models based on the predicted crack propagation and self-healing includes the following steps: S41. Measure the data of cracks in real time within the set measurement area; Real-time acquisition of crack image data within the measurement area in multiple frames, and obtaining fused crack image data through multi-frame image fusion; Crack data within the measurement area is calculated based on the scale of the fused crack image data and the actual distance; S42. After the measurement is completed, predict the crack propagation and self-healing of cracks based on the UHPC data after creep prediction, and construct crack propagation and self-healing models based on the predicted crack propagation and self-healing of cracks.

[0013] Preferably, the process of predicting crack propagation and self-healing based on the UHPC data after creep prediction after the measurement is completed, and constructing crack propagation and self-healing models based on the predicted crack propagation and self-healing, includes the following steps: Cracks in UHPC samples are prefabricated proportionally based on crack data within the measurement area, and crack propagation and self-healing are predicted based on the prefabricated UHPC samples. The predicted crack propagation is shown below: ; in, Indicates the flexural strength of the UHPC sample. This indicates the peak stress experienced by the UHPC sample. The length of the UHPC sample. These represent the height and width of the UHPC sample, respectively. This indicates the depth of the precast crack in the UHPC sample; The bending strength under the peak stress of the UHPC sample is set as the fracture threshold. When the stress on the selected UHPC sample exceeds the peak stress, the crack enters the unstable propagation stage; otherwise, the crack does not propagate. The predicted self-healing scenario for the crack is shown below: Under standard conditions, when no organic healing agent is used, three main types of healing reactions may occur during the self-healing process of UHPC: the first is the secondary hydration of unhydrated cement particles, the second is the active component in the incorporated mineral materials. and The third type of chemical reaction is The formation of precipitate; Models for crack propagation and self-healing are constructed based on predicted crack propagation and self-healing outcomes. Set the crack propagation model as ,in, To select the scale between the sample and the actual measurement area, This represents a crack propagation model; Parameters of three main healing responses during the self-healing process of UHPC were collected, and the healing rates of the three main healing responses were calculated by the controlled variable method and Faraday effect. A self-healing model was constructed based on the scale and the healing rates of the three main types of healing responses. , This represents a self-healing model. Indicates the healing rate.

[0014] Preferably, the coordinated control of cracks within the measurement area of ​​the ultra-large span hybrid structure based on constructing a crack propagation and self-healing model includes the following steps: The UHPC data after creep prediction is input into the crack propagation model to determine the crack propagation situation at the next moment. When the crack enters the unstable propagation stage, the crack healing rate can be improved by using organic healing agent materials. If the crack does not propagate in the next moment, the crack self-healing parameters are calculated using the crack self-healing model.

[0015] This invention also provides a UHPC creep prediction and crack co-control system suitable for ultra-large span hybrid structures, which is used to realize a UHPC creep prediction and crack co-control method suitable for ultra-large span hybrid structures. The system includes: a data acquisition module, a data processing module, a creep prediction module, a crack propagation module, a crack self-healing module, and a co-control module. The data acquisition module is used to collect creep data and environmental data of the UHPC; The data processing module is used to process the collected UHPC creep data to obtain processed UHPC creep data. The creep prediction module is used to perform creep prediction on the processed UHPC creep data to obtain UHPC data after creep prediction. The crack propagation module is used to collect crack data and predict crack propagation based on UHPC data after creep prediction, and at the same time construct a crack propagation model based on the predicted crack propagation. The crack self-healing module is used to predict the crack self-healing situation based on the UHPC data after creep prediction, and to construct a crack self-healing model based on the prediction. The collaborative control module is used to collaboratively control cracks in the measurement area of ​​the ultra-large span hybrid structure based on crack propagation and crack self-healing models.

[0016] The beneficial effects of this invention are as follows: (1) This invention selects UHPC samples of various ages and collects UHPC creep data using a creep meter based on the selected UHPC samples. At the same time, the collected UHPC creep data is processed by data processing method. After processing, the UHPC creep data is predicted by data analysis method. Then, the measurement area of ​​the ultra-large span hybrid structure is set, and the crack data is measured in real time within the set measurement area. After the measurement is completed, the crack propagation and self-healing of cracks are predicted based on the UHPC data after creep prediction. At the same time, a prediction model is constructed based on the predicted crack propagation and self-healing of cracks. Finally, the cracks in the measurement area of ​​the ultra-large span hybrid structure are controlled in a coordinated manner based on the constructed prediction model, which improves the real-time performance of crack control.

[0017] (2) The present invention analyzes the processed UHPC creep data through data analysis methods to determine the creep coefficient. After determining the creep coefficient, the processed UHPC creep data is predicted by neural network prediction, which improves the accuracy and intelligence of creep prediction.

[0018] (3) This invention improves the accuracy and rationality of crack control by collecting crack image data in multiple frames within the measurement area and monitoring and managing the crack image data from two aspects: predicting crack propagation and crack self-healing. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the UHPC creep prediction and crack synergistic control method of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In a specific embodiment of the present invention, Reference Figure 1As shown, this invention provides a method for co-controlling creep prediction and cracking in UHPCs applicable to ultra-large span hybrid structures, comprising: S1. Select UHPC samples of different ages, and collect UHPC creep data using a creep meter based on the selected UHPC samples. S2. The collected UHPC creep data is processed using data processing methods to obtain the processed UHPC creep data; S3. Based on the processed UHPC creep data, creep prediction is performed on the processed UHPC creep data using data analysis methods to obtain the UHPC data after creep prediction. S4. Set the measurement area of ​​the ultra-large span hybrid structure, and measure the crack data in real time within the set measurement area. After the measurement is completed, predict the crack propagation and self-healing based on the UHPC data after creep prediction, and construct crack propagation and self-healing models based on the predicted crack propagation and self-healing. S5. Based on the construction of crack propagation and crack self-healing models, collaborative control of cracks in the measurement area of ​​ultra-large span hybrid structures is carried out. Furthermore, referring to Figure 1 As shown, selecting UHPC samples of various ages and collecting UHPC creep data using a creep meter based on the selected UHPC samples includes the following steps: Set the selected UHPC samples to be of the same size, place a strain gauge at the axis of the selected UHPC sample, and set multiple measurement points around the strain gauge. Furthermore, ambient temperature, ambient humidity, and initial internal stress of selected UHPC samples were collected, and the initial internal stress data of selected UHPC samples under different ambient temperatures and humidity were recorded using the controlled variable method. Furthermore, the selected UHPC samples were numbered, and then creep experiments were conducted on the selected UHPC samples after the numbering was completed. During the experiment, the selected UHPC sample was measured by the corresponding creep meter installed at the set measurement point. The creep data of UHPC was obtained by applying external stress to the selected UHPC sample through the creep meter. Three creep specimens were stacked on each creep meter, and a ball joint support was placed on the top of each stack of specimens to ensure that the UHPC sample was subjected to axial force. Furthermore, referring to Figure 1 As shown, the collected UHPC creep data is processed using data processing methods to obtain the processed UHPC creep data, including the following steps: The collected UHPC creep data were time-series arranged based on the measurement time of the creep meter, and the UHPC creep data sequence was obtained after the arrangement was completed. Furthermore, the UHPC creep data sequence is traversed sequentially through the data traversal method to remove missing data in the UHPC creep data sequence; Furthermore, missing data in the UHPC creep data sequence are located, the average value of the two adjacent sets of UHPC creep data for the missing data is calculated, and the missing data in the UHPC creep data sequence is filled in based on the calculated average value to obtain the processed UHPC creep data. Furthermore, the steps for traversing the UHPC creep data sequence are as follows: The collected UHPC creep data sequences are traversed to locate UHPC creep data sequences with missing content. After the localization is completed, the collected UHPC creep data sequence is traversed. If duplicate data is detected in the UHPC creep data during the traversal, the duplicate UHPC creep data is deleted after the traversal. Furthermore, referring to Figure 1 As shown, based on the processed UHPC creep data, creep prediction is performed on the processed UHPC creep data using data analysis methods to obtain the creep-predicted UHPC data, including the following steps: S31. Based on the processed UHPC creep data, the processed UHPC creep data is analyzed using data analysis methods to determine the creep coefficient; The processed UHPC creep data were fitted using the linear least squares method, and the creep coefficient of the corresponding medium inside the refrigeration unit was calculated based on the fitting results. Based on the processed UHPC creep data, the coordinates of the UHPC creep data are set as follows: ; in, This represents the first set of processed UHPC creep data and the corresponding time. Indicates the first UHPC creep data after group processing and corresponding time; The formula for the linear relationship between two adjacent sets of coordinates is defined as follows: ; The linear least squares fitting formula is shown below: ; in, This represents the fitted UHPC creep data. Represents the slope parameter. Represents linear parameters; The slope parameter is set to the creep coefficient of the processed UHPC creep data; S32. Based on the determined creep coefficient, creep prediction is performed on the processed UHPC creep data using data prediction methods to obtain the UHPC data after creep prediction. S321. Determine the creep prediction model based on the determined creep coefficient; During the process of applying external stress to the selected UHPC sample by the creep meter, a creep prediction model is set up and determined based on the determined creep coefficient. The creep prediction model is set as follows: ; in, Indicates UHPC samples in The measured elastic modulus, These represent different ages of the UHPC samples. This represents the fundamental creep caused by unit stress during the UHPC sample measurement process. This represents the drying creep caused by unit stress during the UHPC sample measurement process. Indicates the error parameter. Represents the creep coefficient during the UHPC sample measurement process; S322. Based on the creep prediction model, creep prediction is performed on the processed UHPC creep data through data prediction to obtain the creep-predicted UHPC data. The processed UHPC creep data and creep prediction model are input into a BP neural network. The structure of the BP neural network includes an input layer, a hidden layer, and an output layer. The prediction process of the BP neural network includes a forward propagation process and an error back-adjustment process. The creep prediction model is set as the training function of a BP neural network; Set the input set of the input layer as ,in Indicates the input number of the first... The output set of the output layer of the UHPC creep data after group processing is as follows: ,in Indicates the output of the first UHPC data after creep prediction; set up The weights from the input layer to the hidden layer; The forward propagation process of the BP neural network is defined as follows: the processed UHPC creep data is input into the hidden layer of the BP neural network through the input layer of the BP neural network. The hidden layer of the BP neural network processes the input processed UHPC creep data based on the training function of the BP neural network and the weights from the input layer to the hidden layer. After the processing is completed, the processing result is transmitted to the output layer of the BP neural network to obtain the UHPC data after creep prediction. Set an error threshold between the output layer and the expected value. When the error between the output layer and the expected value is greater than or equal to the set error threshold, adjust the weights from the hidden layer to the output layer through the error back-adjustment process and perform iterative calculations until the error converges. Output the corresponding BP neural network model and the UHPC data after the creep prediction at the next time step. Furthermore, referring to Figure 1 As shown, a measurement area for a super-large span hybrid structure is defined, and crack data is measured in real time within the defined measurement area. After the measurement is completed, the crack propagation and self-healing are predicted based on the UHPC data after creep prediction. Based on the predicted crack propagation and self-healing, a crack propagation and self-healing model is constructed, including the following steps: S41. Measure the data of cracks in real time within the set measurement area; Real-time acquisition of crack image data within the measurement area in multiple frames, and obtaining fused crack image data through multi-frame image fusion; Furthermore, crack data within the measurement area is calculated based on the scale of the fused crack image data and the actual distance; S42. After the measurement is completed, predict the crack propagation and self-healing of cracks based on the UHPC data after creep prediction, and construct crack propagation and self-healing models based on the predicted crack propagation and self-healing of cracks. Cracks in UHPC samples are prefabricated proportionally based on crack data within the measurement area, and crack propagation and self-healing are predicted based on the prefabricated UHPC samples. The predicted crack propagation is shown below: ; in, Indicates the flexural strength of the UHPC sample. This indicates the peak stress experienced by the UHPC sample. The length of the UHPC sample. These represent the height and width of the UHPC sample, respectively. This indicates the depth of the precast crack in the UHPC sample; Furthermore, the bending strength under the peak stress of the UHPC sample is set as the fracture threshold. When the stress on the selected UHPC sample exceeds the peak stress, the crack enters the unstable propagation stage; otherwise, the crack does not propagate. The predicted self-healing scenario for the crack is shown below: Under standard conditions, when no organic healing agent is used, three main types of healing reactions may occur during the self-healing process of UHPC: the first is the secondary hydration of unhydrated cement particles, the second is the active component in the incorporated mineral materials. and The third type of chemical reaction is The formation of precipitate; Furthermore, crack propagation and self-healing models are constructed based on the predicted crack propagation and self-healing outcomes. Set the crack propagation model as ,in, To select the scale between the sample and the actual measurement area, This represents a crack propagation model; Furthermore, parameters of the three main healing responses during the self-healing process of UHPC were collected, and the healing rates of the three main healing responses were calculated by the controlled variable method and Faraday effect. When the ambient temperature and humidity are constant, the formulas for calculating the healing rate of the three main healing responses are as follows: ; in, Indicates the healing rate. Indicates Faraday efficiency. This indicates the amount of mineral molecules transferred during the self-healing process of the UHPC sample. Denotes Faraday's constant. Indicates the amount of mineral molecules transferred corresponding to the j-th type of healing reaction during the self-healing process of the UHPC sample; Furthermore, a self-healing model was constructed based on the scale and the healing rates of the three main types of healing responses. , This represents a self-healing model; Furthermore, referring to Figure 1 As shown, the collaborative control of cracks within the measurement area of ​​a super-large span hybrid structure based on constructing crack propagation and self-healing models includes the following steps: The UHPC data after creep prediction is input into the crack propagation model to determine the crack propagation situation at the next moment. When the crack enters the unstable propagation stage, the crack healing rate can be improved by using organic healing agent materials. If the crack does not propagate in the next moment, the crack self-healing parameters are calculated using the crack self-healing model. In one specific embodiment, the UHPC creep prediction and crack co-control system applicable to ultra-large span hybrid structures is used to implement a UHPC creep prediction and crack co-control method applicable to ultra-large span hybrid structures. The system includes: a data acquisition module, a data processing module, a creep prediction module, a crack propagation module, a crack self-healing module, and a co-control module. The data acquisition module is used to collect creep data and environmental data of the UHPC; The data processing module is used to process the collected UHPC creep data to obtain processed UHPC creep data. The creep prediction module is used to perform creep prediction on the processed UHPC creep data to obtain UHPC data after creep prediction. The crack propagation module is used to collect crack data and predict crack propagation based on UHPC data after creep prediction, and at the same time construct a crack propagation model based on the predicted crack propagation. The crack self-healing module is used to predict the crack self-healing situation based on the UHPC data after creep prediction, and to construct a crack self-healing model based on the prediction. The collaborative control module is used to collaboratively control cracks in the measurement area of ​​the ultra-large span hybrid structure based on crack propagation and crack self-healing models.

[0023] It should be noted that, The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for co-controlling creep prediction and crack control in UHPC (Ultra-High-Performance Computation) structures with ultra-large spans, characterized in that, Includes the following steps: S1. Select UHPC samples of different ages, and collect UHPC creep data using a creep meter based on the selected UHPC samples. S2. The collected UHPC creep data is processed using data processing methods to obtain the processed UHPC creep data; S3. Based on the processed UHPC creep data, creep prediction is performed on the processed UHPC creep data using data analysis methods to obtain the UHPC data after creep prediction. S4. Set the measurement area of ​​the ultra-large span hybrid structure, and measure the crack data in real time within the set measurement area. After the measurement is completed, predict the crack propagation and self-healing based on the UHPC data after creep prediction, and construct crack propagation and self-healing models based on the predicted crack propagation and self-healing. S5. Based on the construction of crack propagation and crack self-healing models, collaborative control of cracks in the measurement area of ​​ultra-large span hybrid structures is carried out.

2. The method for co-controlling creep prediction and crack control of UHPC applicable to ultra-large span hybrid structures according to claim 1, characterized in that, The process of selecting UHPC samples at various ages and collecting UHPC creep data using a creep meter based on the selected UHPC samples includes the following steps: Set the selected UHPC samples to be of the same size, place a strain gauge at the axis of the selected UHPC sample, and set multiple measurement points around the strain gauge. Ambient temperature, ambient humidity, and initial internal stress of selected UHPC samples were collected. The initial internal stress data of selected UHPC samples under different ambient temperatures and humidity were recorded by controlling the variable method. The selected UHPC samples were numbered, and then creep experiments were performed on the selected UHPC samples after the numbering was completed. During the experiment, the selected UHPC sample was measured by the corresponding creep meter installed at the set measurement point. The creep data of UHPC was obtained by applying external stress to the selected UHPC sample through the creep meter. Three creep specimens were stacked on each creep meter, and a ball joint support was placed on the top of each stack of specimens to ensure that the UHPC specimens were subjected to axial force.

3. The method for co-controlling creep prediction and crack control of UHPC applicable to ultra-large span hybrid structures according to claim 1, characterized in that, The process of processing the collected UHPC creep data to obtain the processed UHPC creep data includes the following steps: The collected UHPC creep data were time-series arranged based on the measurement time of the creep meter, and the UHPC creep data sequence was obtained after the arrangement was completed. The UHPC creep data sequence is traversed sequentially by data traversal to remove missing data in the UHPC creep data sequence; The missing data in the UHPC creep data sequence is located, the average value of the two adjacent UHPC creep data is calculated, and the missing data in the UHPC creep data sequence is filled in based on the calculated average value to obtain the processed UHPC creep data.

4. The method for co-controlling creep prediction and crack control of UHPC applicable to ultra-large span hybrid structures according to claim 1, characterized in that, The process of predicting creep based on the processed UHPC creep data using data analysis methods to obtain the predicted UHPC data includes the following steps: S31. Based on the processed UHPC creep data, the processed UHPC creep data is analyzed using data analysis methods to determine the creep coefficient; The processed UHPC creep data were fitted using the linear least squares method, and the creep coefficient of the corresponding medium inside the refrigeration unit was calculated based on the fitting results. Based on the processed UHPC creep data, the coordinates of the UHPC creep data are set as follows: ; in, This represents the first set of processed UHPC creep data and the corresponding time. Indicates the first UHPC creep data after group processing and corresponding time; The formula for the linear relationship between two adjacent sets of coordinates is defined as follows: ; The linear least squares fitting formula is shown below: ； in, This represents the fitted UHPC creep data. Represents the slope parameter. Represents linear parameters; The slope parameter is set to the creep coefficient of the processed UHPC creep data; S32. Based on the determined creep coefficient, creep prediction is performed on the processed UHPC creep data using data prediction methods to obtain the creep-predicted UHPC data.

5. The method for co-controlling creep prediction and crack control of UHPC applicable to ultra-large span hybrid structures according to claim 4, characterized in that, The process of predicting creep in the processed UHPC creep data based on a determined creep coefficient, and obtaining the UHPC data after creep prediction, includes the following steps: S321. Determine the creep prediction model based on the determined creep coefficient; During the process of applying external stress to the selected UHPC sample by the creep meter, a creep prediction model is set up and determined based on the determined creep coefficient. The creep prediction model is set as follows: ； in, Indicates UHPC samples in The measured elastic modulus, These represent different ages of the UHPC samples. This represents the fundamental creep caused by unit stress during the UHPC sample measurement process. This represents the drying creep caused by unit stress during the UHPC sample measurement process. Indicates the error parameter. Represents the creep coefficient during the UHPC sample measurement process; S322. Based on the creep prediction model, creep prediction is performed on the processed UHPC creep data through data prediction to obtain the creep-predicted UHPC data.

6. The method for co-controlling creep prediction and crack control of UHPC applicable to ultra-large span hybrid structures according to claim 5, characterized in that, The process of predicting creep in the processed UHPC creep data based on the creep prediction model, and obtaining the creep-predicted UHPC data, includes the following steps: The processed UHPC creep data and creep prediction model are input into a BP neural network. The structure of the BP neural network includes an input layer, a hidden layer, and an output layer. The prediction process of the BP neural network includes a forward propagation process and an error back-adjustment process. The creep prediction model is set as the training function of a BP neural network; set up The weights from the input layer to the hidden layer; The forward propagation process of the BP neural network is defined as follows: the processed UHPC creep data is input into the hidden layer of the BP neural network through the input layer of the BP neural network. The hidden layer of the BP neural network processes the input processed UHPC creep data based on the training function of the BP neural network and the weights from the input layer to the hidden layer. After the processing is completed, the processing result is transmitted to the output layer of the BP neural network to obtain the UHPC data after creep prediction. Set an error threshold between the output layer and the expected value. When the error between the output layer and the expected value is greater than or equal to the set error threshold, adjust the weights from the hidden layer to the output layer through the error back-adjustment process and perform iterative calculations until the error converges. Output the corresponding BP neural network model and the UHPC data after creep prediction at the next time step.

7. The method for co-controlling creep prediction and crack control of UHPC applicable to ultra-large span hybrid structures according to claim 1, characterized in that, The process of setting up a measurement area for the ultra-large span hybrid structure, measuring crack data in real time within the set measurement area, predicting crack propagation and self-healing based on the UHPC data after creep prediction, and constructing crack propagation and self-healing models based on the predicted crack propagation and self-healing includes the following steps: S41. Measure the data of cracks in real time within the set measurement area; Real-time acquisition of crack image data within the measurement area in multiple frames, and obtaining fused crack image data through multi-frame image fusion; Crack data within the measurement area is calculated based on the scale of the fused crack image data and the actual distance; S42. After the measurement is completed, predict the crack propagation and self-healing of cracks based on the UHPC data after creep prediction, and construct crack propagation and self-healing models based on the predicted crack propagation and self-healing of cracks.

8. The method for co-controlling creep prediction and crack control of UHPC in ultra-large span hybrid structures according to claim 7, characterized in that, After the measurement is completed, the crack propagation and self-healing conditions are predicted based on the UHPC data after creep prediction. The crack propagation and self-healing models are then constructed based on the predicted crack propagation and self-healing conditions, including the following steps: Cracks in UHPC samples are prefabricated proportionally based on crack data within the measurement area, and crack propagation and self-healing are predicted based on the prefabricated UHPC samples. The predicted crack propagation is shown below: ； in, Indicates the flexural strength of the UHPC sample. This indicates the peak stress experienced by the UHPC sample. The length of the UHPC sample. These represent the height and width of the UHPC sample, respectively. This indicates the depth of the precast crack in the UHPC sample; The bending strength under the peak stress of the UHPC sample is set as the fracture threshold. When the stress on the selected UHPC sample exceeds the peak stress, the crack enters the unstable propagation stage; otherwise, the crack does not propagate. The predicted self-healing scenario for the crack is shown below: Under standard conditions, when no organic healing agent is used, three main types of healing reactions may occur during the self-healing process of UHPC: the first is the secondary hydration of unhydrated cement particles, the second is the active component in the incorporated mineral materials. and The third type of chemical reaction is The formation of precipitate; Models for crack propagation and self-healing are constructed based on predicted crack propagation and self-healing outcomes. Set the crack propagation model as ,in, To select the scale between the sample and the actual measurement area, This represents a crack propagation model; Parameters of three main healing responses during the self-healing process of UHPC were collected, and the healing rates of the three main healing responses were calculated by the controlled variable method and Faraday effect. A self-healing model was constructed based on the scale and the healing rates of the three main types of healing responses. , This represents a self-healing model. Indicates the healing rate.

9. The method for co-controlling creep prediction and crack control of UHPC applicable to ultra-large span hybrid structures according to claim 1, characterized in that, The method for collaboratively controlling cracks within the measurement area of ​​a super-large span hybrid structure based on constructing crack propagation and self-healing models includes the following steps: The UHPC data after creep prediction is input into the crack propagation model to determine the crack propagation situation at the next moment. When the crack enters the unstable propagation stage, the crack healing rate can be improved by using organic healing agent materials. If the crack does not propagate in the next moment, the crack self-healing parameters are calculated using the crack self-healing model.

10. A method for co-controlling creep prediction and cracking in UHPC of claim 1, applicable to ultra-large span hybrid structures, characterized in that, Also includes: The system includes a data acquisition module, a data processing module, a creep prediction module, a crack propagation module, a crack self-healing module, and a collaborative control module. The data acquisition module is used to collect creep data and environmental data of the UHPC; The data processing module is used to process the collected UHPC creep data to obtain processed UHPC creep data. The creep prediction module is used to perform creep prediction on the processed UHPC creep data to obtain UHPC data after creep prediction. The crack propagation module is used to collect crack data and predict crack propagation based on UHPC data after creep prediction, and at the same time construct a crack propagation model based on the predicted crack propagation. The crack self-healing module is used to predict the crack self-healing situation based on the UHPC data after creep prediction, and to construct a crack self-healing model based on the prediction. The collaborative control module is used to collaboratively control cracks in the measurement area of ​​the ultra-large span hybrid structure based on crack propagation and crack self-healing models.

Images