A data-driven old bridge demolition construction monitoring method

Abstract

The application discloses a kind of old bridge demolition construction monitoring methods based on data driving, from the angle of the structural state evolution of bridge whole life, through the continuous analysis of bridge design-new construction-operation-demolition and so on Key stage Real state of old bridge structure is difficult to determine, to the concept of data-driven quality, by establishing the corresponding automatic monitoring system of structure and old bridge demolition construction control auxiliary decision mechanism, solve the key problems such as old bridge parameter identification and behavior prediction, simultaneously realize the deviation optimal control in the process of bridge demolition by optimization theory.

Description

Technical Field

[0001] This invention relates to the field of bridge construction technology. More specifically, this invention relates to a data-driven method for monitoring the demolition of old bridges. Background Technology

[0002] Some bridges built in my country in the 1990s, due to factors such as imperfect survey and design standards, low construction levels, overloading after service, and inadequate maintenance, face the situation of having to be demolished and rebuilt after more than 30 years of service due to serious defects. At the same time, according to relevant Chinese bridge design specifications, bridges are designed with a lifespan of 50 or 100 years. After reaching the design period, bridges all face the need for reinforcement or demolition and reconstruction. Therefore, it is necessary to study the demolition construction of old bridges.

[0003] Bridge construction monitoring, as a crucial means of ensuring safety and quality during bridge construction, has gained widespread recognition within the industry. Currently, most research on bridge construction monitoring focuses on newly constructed bridges, with less attention paid to the demolition of existing bridges. Existing monitoring of old bridge demolition is often conducted using the same approach as for new bridge construction. Compared to new bridges, old bridges, having undergone operation and maintenance, exhibit significant discrepancies between key parameters reflecting their structural mechanical state and theoretical values. Furthermore, the monitoring process involves substantial data collection and analysis, which traditional construction monitoring theories struggle to handle. Therefore, it is necessary to develop a construction monitoring method suitable for guiding the demolition of old bridges. Summary of the Invention

[0004] One object of the present invention is to solve at least the above-mentioned problems and to provide at least the advantages that will be described later.

[0005] Another objective of this invention is to provide a data-driven method for monitoring the demolition of old bridges, in order to solve the technical problem that existing monitoring schemes for the demolition of old bridges differ significantly from the actual condition of the old bridges, which is detrimental to ensuring the safety of the demolition process.

[0006] To achieve these objectives and other advantages according to the present invention, a data-driven method for monitoring the demolition of old bridges is provided, comprising the following steps:

[0007] S1. Starting from the structural state evolution of the bridge throughout its entire life cycle, the entire life cycle of a bridge includes four key stages: bridge design, new construction, operation, and demolition of the old bridge. Based on the bridge design and new construction plan, a theoretical bridge completion state model is established.

[0008] S2. Monitor the actual condition of the bridge during the operation phase, revise the theoretical bridge completion state model, and obtain the actual state model of the old bridge before demolition.

[0009] S3. Set up a construction plan for the demolition of the old bridge, and obtain a theoretical analysis model for the demolition of the old bridge by combining the actual state model of the old bridge before demolition.

[0010] S4. Based on the structural characteristics and mechanical properties of the bridge to be demolished, determine the structural responses that need attention, divide the old bridge structure into substructures, and determine the key structural parameters that affect the structural response in all substructures through parameter sensitivity analysis. Based on the theoretical analysis model of old bridge demolition, train and establish a substructure parameter identification model and a substructure behavior prediction model. The substructure parameter identification model takes the structural response as input and the structural parameters as output, while the substructure behavior prediction model takes the structural parameters as input and the structural response as output.

[0011] S5. Establish an automated monitoring system for substructure response to monitor the structural response data of substructures;

[0012] S6. Establish an auxiliary decision-making mechanism for the demolition of the old bridge. During the current demolition phase, determine the structural parameters that actually affect the structural response of the current substructure through a substructure parameter identification model. Then, predict and analyze the predicted values ​​of the structural response in subsequent demolition phases using a substructure behavior prediction model. If the predicted values ​​affect the structural construction safety, adjust the corresponding construction measures for the demolition of the old bridge to ensure the structural safety during the demolition phase.

[0013] Preferably, in step S2, the parameters to be corrected include material strength and modulus of elasticity, effective prestress of prestressed steel strands, anchorage length and transfer length of prestressed steel strands, actual geometric dimensions of the structure, stiffness and weight of concrete box girder, and structural damage status.

[0014] Preferably, in step S4, after determining the structural responses to be of interest, each substructure is taken as the research object, and a simulation analysis is performed using the old bridge demolition theoretical analysis model to obtain learning samples. A training set is obtained through orthogonal experiments, and a substructure parameter identification model and a substructure behavior prediction model are obtained through training.

[0015] Preferably, before training, the structural parameters to be identified or the structural responses to be predicted are taken at ±10% of the theoretical values. Simulation analysis is performed using the old bridge demolition theoretical analysis model to obtain the corresponding changes in structural responses or structural parameters. After normalization, training samples are obtained. The training set is obtained through orthogonal experiments. A Bayes network is used to establish a substructure parameter identification model and a substructure behavior prediction model.

[0016] Preferably, in step S5, the substructure response automated monitoring system includes a sensor detection group, a data acquisition and transmission module, and a cloud server. Each sensor detection group is set up for each structural response determined in step S4. Each group of sensors transmits the monitoring data to the cloud server via wireless communication through the data acquisition and transmission module. The cloud server then establishes a database of measured values ​​for each substructure.

[0017] Preferably, in step S6, an auxiliary decision-making system for the demolition of an old bridge is set up, which is connected to the measured value database. The auxiliary decision-making system for the demolition of an old bridge includes an optimization analysis module. The auxiliary decision-making system for the demolition of an old bridge retrieves the monitoring data stored in the measured value database. Based on the substructure parameter identification model and the substructure behavior prediction model, the system identifies the true key parameters through the response error of the substructure parameter identification model. The system then substitutes the true key parameters into the substructure behavior prediction model to obtain the predicted values ​​of the structural response in the subsequent construction stage. The system determines whether the predicted values ​​affect the structural construction safety. If they do, the optimization analysis module uses the substructure behavior prediction model to determine the optimal adjustment range of the parameters and adjusts the construction measures corresponding to the demolition of the old bridge according to the optimal adjustment range of the parameters.

[0018] Preferably, the construction measures adjusted in step S6 include adjusting the beam length of the demolished beam segment, local reinforcement, adjusting the timing of prestressed removal, and correcting the self-anchoring effect.

[0019] The present invention has at least the following beneficial effects: The data-driven monitoring method for the demolition of old bridges of the present invention starts from the perspective of the structural state evolution of the bridge throughout its entire life cycle. It solves the problem of difficulty in determining the true state of the old bridge structure by continuously analyzing key stages such as bridge design, new construction, operation and demolition. With the concept of data-driven quality, it solves key problems such as old bridge parameter identification and behavior prediction by establishing a corresponding automated monitoring system for the structure and an auxiliary decision-making mechanism for the control of old bridge demolition construction. At the same time, it achieves optimal control of deviations during the bridge demolition process through optimization theory.

[0020] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of a finite element model of a prestressed concrete continuous rigid frame bridge according to design drawings, which is an embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram of the selection of correction parameters for the finite element model of a prestressed concrete continuous rigid frame bridge according to an embodiment of the present invention.

[0023] Figure 3 This is a schematic diagram of the bridge substructure division according to an embodiment of the present invention;

[0024] Figure 4 A flowchart illustrating the establishment of a substructure parameter identification model and a substructure behavior prediction model according to an embodiment of the present invention;

[0025] Figure 5 A flowchart illustrating an automated monitoring system for substructure response according to an embodiment of the present invention;

[0026] Figure 6 This is a flowchart of step S6 in one embodiment of the present invention;

[0027] Figure 7 This is a construction control flowchart for each construction stage during the bridge demolition process according to an embodiment of the present invention. Detailed Implementation

[0028] The present invention will now be described in further detail with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0029] It should be noted that, unless otherwise specified, the experimental methods described in the following embodiments are all conventional methods, and the reagents and materials described are all commercially available unless otherwise specified. In the description of this invention, the terms "lateral", "longitudinal", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", and "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0030] like Figure 1-7 As shown, this invention provides a data-driven method for monitoring the demolition of old bridges, comprising the following steps:

[0031] S1. Starting from the structural state evolution of the bridge throughout its entire life cycle, the entire life cycle of a bridge includes four key stages: bridge design, new construction, operation, and demolition of the old bridge. Based on the bridge design and new construction plan, a theoretical bridge completion state model is established.

[0032] When establishing the theoretical completed bridge state model, the bridge design finite element model is first established using midas software based on the design drawings. Then, the theoretical completed bridge state model is obtained based on the bridge design finite element model in conjunction with the new design scheme.

[0033] S2. Monitor the actual condition of the bridge during the operation phase, revise the theoretical bridge completion state model, and obtain the actual state model of the old bridge before demolition.

[0034] The theoretical bridge completion state model does not consider the impact of construction errors on the bridge completion state. Therefore, the theoretical bridge completion state model is modified by combining bridge handover inspection, reinforcement and maintenance, load test and construction monitoring to obtain the actual bridge completion state model.

[0035] S3. Set up a construction plan for the demolition of the old bridge, and obtain a theoretical analysis model for the demolition of the old bridge by combining the actual state model of the old bridge before demolition.

[0036] S4. Based on the structural characteristics and mechanical properties of the bridge to be demolished, determine the structural responses that need attention, divide the old bridge structure into substructures, and determine the key structural parameters that affect the structural response in all substructures through parameter sensitivity analysis. Based on the theoretical analysis model of old bridge demolition, train and establish a substructure parameter identification model and a substructure behavior prediction model. The substructure parameter identification model takes the structural response as input and the structural parameters as output, while the substructure behavior prediction model takes the structural parameters as input and the structural response as output.

[0037] Structural response includes parameters such as stress, alignment, cable force, and prestress. Components of a bridge constructed in the same batch or under similar conditions constitute the same substructure. The substructure parameter identification model and the substructure behavior prediction model are reverse processes. By setting up the substructure parameter identification model, a certain structural response is input, and the corresponding structural parameters are output to verify and identify which parameters will significantly affect the structural response in the current stage of bridge demolition construction. Thus, the parameters that significantly affect these structural responses are identified as key structural parameters, completing the identification process. As for the substructure behavior prediction model, the structural parameters are input to obtain the predicted values ​​of the structural response in subsequent construction stages.

[0038] S5. Establish an automated monitoring system for substructure response to monitor the structural response data of the substructure. Based on the structural response determined in the previous step, set up the monitoring system to acquire monitoring data in real time and establish a database. This database will serve as a learning sample during the model building process and will also be used to compare the predicted values ​​of the structural response in the subsequent construction phase. If the difference is too large, it indicates that the demolition construction plan needs to be modified to ensure construction safety.

[0039] S6. Establish an auxiliary decision-making mechanism for the demolition of the old bridge. During the current demolition construction phase, the structural parameters that actually affect the structural response of the current substructure are determined by the substructure parameter identification model. Then, the predicted values ​​of the structural response in the subsequent demolition construction phase are predicted and analyzed by the substructure behavior prediction model. If the predicted values ​​affect the structural construction safety, the corresponding construction measures for the demolition of the old bridge are adjusted to ensure the structural safety during the demolition phase.

[0040] The decision-making support mechanism for the demolition of old bridges depends on the comparative structure of data. To determine whether the predicted values ​​will affect the safety of structural construction, an allowable error range can be set for the data. If the data exceeds the allowable error range, it indicates that the current demolition method will have an adverse impact on the bridge and the surrounding environment. For specific structural parameters that exceed the allowable error range, corresponding adjustment measures are set to adjust the current structural parameters to a safe range from the perspective of prediction. Feedback is then provided to carry out actual construction work. Real-time monitoring and feedback ensure the smooth progress of each stage of the old bridge demolition.

[0041] This invention presents a data-driven monitoring method for the demolition of old bridges. Starting from the perspective of the structural state evolution throughout the bridge's entire lifespan, it continuously analyzes key stages such as bridge design, new construction, operation, and demolition. This process ultimately yields a theoretical analysis model for the bridge demolition construction phase. The model is established by iteratively creating a structural model for the bridge's current lifespan and then using measured data from that period for correction. The final model most closely approximates the actual state of the bridge before demolition. Based on this theoretical analysis model, a substructure parameter identification model and a substructure behavior prediction model are established. One model identifies sensitive parameters affecting the bridge's structural response, while the other serves as a data model for predicting subsequent structural construction based on these identified parameters. This identifies key parameters reflecting the bridge's structural mechanical state, avoiding the large and complex data analysis required by traditional methods. Then, during the demolition phase, a targeted automated monitoring system for bridge stress and alignment is established. The substructure parameter identification model and substructure behavior prediction model are combined with the automated monitoring system to create an auxiliary decision-making mechanism for the demolition construction, guiding each stage of the demolition process.

[0042] In another technical solution, such as Figure 2 As shown, in step S2, the parameters that need to be corrected include material strength and modulus of elasticity, effective prestress of prestressed steel strands, anchorage length and transfer length of prestressed steel strands, actual geometric dimensions of the structure, stiffness and weight of the concrete box girder, and structural damage status. These parameters can be set according to the important parameters reflecting the structural safety status of different bridges.

[0043] In another technical solution, such as Figure 3 As shown, in step S4, after determining the structural response that needs to be focused on, each substructure is taken as the research object, and the old bridge demolition theory analysis model is used to conduct simulation analysis to obtain learning samples. The training set is obtained through orthogonal experiments, and the substructure parameter identification model and substructure behavior prediction model are obtained through training.

[0044] Based on the theoretical analysis model of the bridge demolition construction stage, the structural responses that need attention are determined according to the structural characteristics and mechanical properties of the bridge to be demolished. By dividing the entire bridge into substructures and conducting orthogonal experiments, the key parameters affecting the stress and deformation of each substructure are obtained. Taking each substructure as the research object, the training set is obtained through orthogonal experiments.

[0045] In another technical solution, before training, the structural parameters to be identified or the structural responses to be predicted are taken at ±10% of the theoretical values. Simulation analysis is performed using the old bridge demolition theoretical analysis model to obtain the corresponding changes in structural responses or structural parameters. After normalization, training samples are obtained. The training set is obtained through orthogonal experiments. A Bayes network is used to establish a substructure parameter identification model and a substructure behavior prediction model, which can handle the complex relationships between a large number of variables and effectively describe the dependencies between variables.

[0046] In another technical solution, such as Figure 5 As shown, in step S5, the substructure response automated monitoring system includes a sensor detection group, a data acquisition and transmission module, and a cloud server. Each sensor detection group is set up for each structural response determined in step S4. Each group of sensors transmits the monitoring data to the cloud server through the data acquisition and transmission module using wireless communication. The cloud server then establishes a database of the corresponding measured values ​​for each substructure.

[0047] In another technical solution, such as Figure 6 As shown, in step S6, an auxiliary decision-making system for the demolition of an old bridge is set up, which is connected to the measured value database. The auxiliary decision-making system for the demolition of an old bridge is equipped with an optimization analysis module. The auxiliary decision-making system for the demolition of an old bridge retrieves the monitoring data stored in the measured value database. Based on the substructure parameter identification model and the substructure behavior prediction model, the system obtains the true key parameters by identifying the response error of the substructure parameter identification model. The system then substitutes the true key parameters into the substructure behavior prediction model to obtain the predicted values ​​of the structural response in the subsequent construction stage. The system determines whether the predicted values ​​affect the structural construction safety. If they do, the optimization analysis module uses the substructure behavior prediction model to determine the optimal adjustment range of the parameters and adjusts the construction measures corresponding to the demolition of the old bridge according to the optimal adjustment range of the parameters.

[0048] The following is a specific example of the demolition of an old bridge: A prestressed concrete continuous rigid frame bridge with a span of (39+72+39)m was completed and opened to traffic in 1997. After two repairs, the main beams were demolished because the load-bearing capacity no longer met the requirements. A design finite element model was established based on the design drawings, such as... Figure 1 As shown.

[0049] Based on the finite element model design, the parameters of the model were corrected according to the measured values ​​of key parameters at different stages of the bridge construction. These parameters mainly included material strength and elastic modulus, effective prestress of the prestressed steel strands, anchorage length and transfer length of the prestressed steel strands, actual geometric dimensions of the structure, stiffness and weight of the concrete box girder, and structural damage status. The final corrected parameters selected before demolition in this project are as follows: Figure 2 As shown, after modification, the actual state model is obtained. Combined with the demolition plan, the theoretical analysis model for the demolition of the old bridge is obtained.

[0050] Based on the theoretical analysis model of old bridge demolition, the structural responses that need attention are determined according to the structural characteristics and mechanical properties of the bridge to be demolished. By dividing the entire bridge into substructures and conducting orthogonal experiments, key parameters affecting the stress and deformation of each substructure are obtained. Taking each substructure as the research object, a training set is obtained through orthogonal experiments. Through training, a substructure parameter identification model with structural response as input and structural parameters as output, and a substructure behavior prediction model with structural parameters as input and structural response as output are established. The process is as follows: Figure 3 As shown.

[0051] This project can be divided into 13 substructures based on the cantilever demolition construction stages, such as... Figure 4 As shown, the approach to establishing a substructure parameter identification model is illustrated. Sensitivity analysis reveals that the weight of the main beam is sensitive to the linearity of substructures ⑥-⑩. Therefore, the main beam stiffness is chosen as the target vector, i.e., the parameter to be identified. The elevations of the five foremost segments (substructures ⑧-⑩) during the cantilever construction of the main beam are selected as the input vectors. A Bayesian network is used to establish the substructure parameter identification model. The results of the theoretical analysis model of the old bridge demolition during the demolition stage are used as training samples. The parameters to be identified are taken at ±10% of the theoretical values ​​to obtain the changes in the main beam elevation during the maximum cantilever stage, thus obtaining the model's training samples.

[0052] Nine values ​​were taken for the design unit weight of the main beam concrete, with a fluctuation range of 10% (i.e., 90%γ-110%γ). The values ​​are detailed in Table 1 below. These values ​​were then substituted into the finite element model to calculate the front-end deflection changes of the first five segments (substructures ⑧-⑩), thus obtaining nine sets of learning samples. Three additional sets of test samples were selected, as shown in Table 2 below. Before learning, the samples were normalized, as shown in Table 3 below. The test samples are shown in Table 4 below. After the sample training was completed, the test samples could be tested to simulate the parameter identification process. The predicted values, measured values, and the errors between them are shown in Table 5 below.

[0053] Table 1 Learning Samples

[0054]

[0055]

[0056] Table 2 Test Samples

[0057]

[0058] Table 3 Normalized learning samples

[0059]

[0060] Table 4. Normalized test samples

[0061]

[0062]

[0063] Table 5 shows the error between predicted and measured values.

[0064]

[0065] An automated monitoring system for substructure response is established. Based on the determined structural response, appropriate sensors are selected and wireless transmission is used. Data is transmitted to a cloud server via a data acquisition and transmission module to establish a database of measured values ​​for each substructure. The process is as follows: Figure 5 As shown.

[0066] like Figure 6 As shown, an auxiliary decision-making system for the demolition of the old bridge was established. This system is based on a Bayesian model for substructure parameter identification and a Bayesian model for substructure behavior prediction. First, the response of each substructure at the current construction stage is analyzed. When the difference between theoretical and measured values ​​exceeds 10%, the actual key parameters of each substructure are obtained through the substructure parameter identification model. These actual key parameters are then input into the substructure behavior prediction model to predict the response of each substructure during subsequent construction. When the predicted values ​​affect the structural construction safety, optimization theory is used to analyze and determine the necessary adjustment measures. These measures include adjusting the beam length of the demolished beam segment, local reinforcement measures, timing of prestressed removal, and correction of the self-anchoring effect. The main beam demolition construction control process is as follows: Figure 7 As shown.

[0067] The demolition process of the bridge was monitored, and the measured values ​​and theoretical values ​​of the deviation of the main beam alignment of each key construction segment are shown in Table 6 below. All of them are within 10%, which shows that the monitoring method and system established by the present invention can effectively guide the demolition of the old bridge.

[0068] Table 6. Statistics of maximum displacement monitoring results at the cantilever end during main beam dismantling / mm

[0069]

[0070]

[0071] In summary, the data-driven monitoring method for the demolition of old bridges of this invention addresses the problem of difficulty in determining the true state of old bridge structures by continuously analyzing key stages such as bridge design, new construction, operation, and demolition, from the perspective of the structural state evolution throughout the entire life cycle of the bridge. Adopting a data-driven quality approach, it solves key challenges such as old bridge parameter identification and behavior prediction by establishing a corresponding automated monitoring system and an auxiliary decision-making mechanism for the control of old bridge demolition construction. Furthermore, it achieves optimal control of deviations during the bridge demolition process through optimization theory.

[0072] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A data-driven method for monitoring the demolition of old bridges, characterized in that, Includes the following steps: S1. Starting from the structural state evolution of the bridge throughout its entire life cycle, the entire life cycle of a bridge includes four key stages: bridge design, new construction, operation, and demolition of the old bridge. Based on the bridge design and new construction plan, a theoretical bridge completion state model is established. S2. Monitor the actual condition of the bridge during the operation phase, and revise the theoretical bridge condition model to obtain the actual condition model before the demolition of the old bridge. The parameters to be revised include material strength and elastic modulus, effective prestress of prestressed steel strands, anchorage length and transfer length of prestressed steel strands, actual geometric dimensions of the structure, stiffness and weight of concrete box girder, and structural damage. S3. Set up a construction plan for the demolition of the old bridge, and obtain a theoretical analysis model for the demolition of the old bridge by combining the actual state model of the old bridge before demolition. S4. Based on the structural characteristics and mechanical properties of the bridge to be demolished, determine the structural responses that need attention, divide the old bridge structure into substructures, and determine the key structural parameters that affect the structural response in all substructures through parameter sensitivity analysis. Based on the theoretical analysis model of old bridge demolition, train and establish a substructure parameter identification model and a substructure behavior prediction model. The substructure parameter identification model takes the structural response as input and the structural parameters as output, while the substructure behavior prediction model takes the structural parameters as input and the structural response as output. S5. Establish an automated monitoring system for substructure response to monitor the structural response data of substructures; The substructure response automated monitoring system includes a sensor detection group, a data acquisition and transmission module, and a cloud server. The sensor detection group is set up for each structural response determined in each step S4. Each group of sensors transmits the monitoring data to the cloud server through the data acquisition and transmission module using wireless communication. The cloud server establishes a database of the corresponding measured values ​​for each substructure. S6. Establish an auxiliary decision-making mechanism for the demolition of the old bridge. An auxiliary decision-making system for the demolition of the old bridge, connected to the measured value database, is set up. The auxiliary decision-making system for the demolition of the old bridge includes an optimization analysis module. The system retrieves monitoring data stored in the measured value database. During the current demolition phase, based on the substructure parameter identification model and the substructure behavior prediction model, the system identifies the structural parameters that actually affect the structural response of the current substructure through the response error identification of the substructure parameter identification model. The system then substitutes these actual key parameters into the substructure behavior prediction model to obtain the predicted values ​​of the structural response in subsequent construction phases. The system analyzes the predicted values ​​of the structural response in subsequent demolition phases to determine whether the predicted values ​​affect the structural construction safety. If they do affect the structural construction safety, the optimization analysis module uses the substructure behavior prediction model to determine the optimal adjustment range of the parameters. Based on the optimal adjustment range of the parameters, the system adjusts the corresponding construction measures for the demolition of the old bridge to ensure the structural safety during the demolition phase.

2. The data-driven method for monitoring the demolition of old bridges as described in claim 1, characterized in that, In step S4, after determining the structural responses that need to be focused on, each substructure is taken as the research object, and simulation analysis is performed using the old bridge demolition theoretical analysis model to obtain learning samples. The training set is obtained through orthogonal experiments, and the substructure parameter identification model and substructure behavior prediction model are obtained through training.

3. The data-driven method for monitoring the demolition of old bridges as described in claim 2, characterized in that, Before training, the structural parameters to be identified or the structural responses to be predicted are taken at ±10% of the theoretical values. Simulation analysis is performed using the old bridge demolition theoretical analysis model to obtain the corresponding changes in structural responses or structural parameters. After normalization, training samples are obtained. The training set is obtained through orthogonal experiments. A Bayes network is used to establish a substructure parameter identification model and a substructure behavior prediction model.

4. The data-driven monitoring method for the demolition of old bridges as described in claim 1, characterized in that, The construction measures adjusted in step S6 include adjusting the beam length of the demolished beam segment, local reinforcement, adjusting the timing of prestressed removal, and correcting the self-anchoring effect.

Images