Safety data identification and early warning method for shallow buried tunnel construction close to tall structures

Abstract

The present invention is a safety data identification and early warning method for high-rise structures near shallow tunnel construction, including safety monitoring equipment, structural modal calibration during construction, structural status modal monitoring, operational modal model calibration, buckling modal analysis model establishment, buckling modal ultimate bearing capacity calculation, dynamic monitoring and early warning, and early warning criteria, aiming to solve the difficulties in implementing underground foundation structure monitoring of existing high-rise structures. The traditional upper structure monitoring method using stress, strain, deformation and displacement as reference quantities has a safety early warning lag deviation and cannot reflect the overall bearing capacity attenuation of the high-rise structure. Starting from the vibration modal change mechanism caused by the attenuation of foundation bearing capacity, the vibration mode of the high-rise structure is used as the safety monitoring object; the buckling stability ultimate bearing capacity at the characteristic order corresponding to the first mutation of strain energy in the buckling modal analysis is used as an indicator to characterize the bearing capacity of the high-rise structure, and the attenuation rate of the bearing capacity indicator is used as a safety early warning indicator.

Description

Technical Field

[0001] The invention belongs to the technical field of tunnel construction, and relates to data identification and recording carrier processing, and in particular to a safety data identification and early warning method for shallow buried tunnel construction close to tall structures. Background Art

[0002] In recent years, with the continuous development of urban subway construction in my country, the number of tunnels has gradually increased. The subway has gradually become the lifeline of urban transportation in China. With the development of urban underground space and subway systems, it is often restricted by existing tall structures, geological conditions and other factors, and it is inevitable that shield construction will be close to existing tall structures. In particular, with the increase in subway lines, tunnel excavation will inevitably cause soil disturbance and stratum loss, which will have a significant impact on the stress and deformation of nearby tall structures. Excessive deformation will cause the tall structures to sink and collapse, which is not conducive to the structural safety of existing operating rail transit tunnels, and thus affects the safe operation of subway lines. For this reason, a safety monitoring and early warning method for shallow buried tunnel construction close to tall structures was developed.

[0003] CN 115288704 A discloses a safety risk assessment and construction method for pile foundation settlement caused by shield tunneling. The invention includes predicting the settlement prediction values ​​of pile foundations at several different positions according to a pre-established finite element numerical model in the current shield construction, and verifying it in combination with the measured settlement value; adjusting the pile-tunnel horizontal distance and the pile foundation length of the pile foundation, predicting the settlement values ​​of the pile foundations at different positions according to the finite element numerical model, and generating a settlement equivalent two-dimensional map with safety risk levels of pile foundations at different positions with reference to the pile foundation settlement safety risk level; and performing shield construction after executing the corresponding protection measures according to the predicted safety risk level of the pile foundation. The present invention verifies the numerical model by summarizing the shield construction parameters, and then generates a two-dimensional map of equivalent values ​​for the pile foundation settlement safety risk level division, so that the safety risk level of the adjacent pile foundation in the subsequent construction is timely and reliable.

[0004] CN 117988869 A discloses a shield adjacent structure safety prediction control method and system based on intelligent algorithm. The invention includes collecting shield construction parameters to obtain sample data, preprocessing the sample data, introducing Bayesian optimization method to optimize the hyperparameters of the natural gradient boosting prediction model, constructing and training the optimized natural gradient boosting prediction model with the output variable as the control target according to the normalized sample data set, predicting the output variable through the optimized natural gradient boosting prediction model, and outputting the prediction result; dividing the shield adjacent structure safety risk level according to the prediction result; outputting the corresponding safety control strategy according to the shield adjacent structure safety risk level, and realizing the intelligent prediction and control of the adjacent structure safety risk during the shield construction process.

[0005] The above patent still uses stress, strain, deformation and displacement as reference quantities in the traditional monitoring method of the superstructure, which has a safety warning lag deviation and cannot reflect the overall bearing capacity attenuation of the tall structure. It is necessary to propose physical quantities that characterize the overall bearing capacity attenuation of the tall structure and construct a method for its detection and estimation. Summary of the invention

[0006] Technical problem: The purpose of the present invention is to solve the problems in the above-mentioned technology by processing data identification and recording carriers, and proposes a safety data identification and early warning method for shallow buried tunnel construction close to tall structures. It overcomes the safety early warning lag deviation of the traditional monitoring method of superstructures using stress, strain, deformation and displacement as reference quantities, and reveals the influence of the attenuation of the bearing capacity of tall structures during shallow buried tunnel construction by measuring the vibration modal changes of tall structures. The attenuation rate of the buckling stability limit bearing capacity is used as a judgment indicator for early warning, avoiding the hysteresis of the traditional monitoring method of superstructures using stress, strain, deformation and displacement as reference quantities.

[0007] Technical solution: The present invention adopts a safety data identification and early warning method for shallow buried tunnel construction close to tall structures to perform data identification and record carrier processing.

[0008] The safety data identification and early warning method for shallow buried tunnel construction close to tall structures includes eight parts, including safety monitoring equipment, structural modal calibration during construction, structural status modal monitoring, operational modal model calibration, buckling modal analysis model establishment, buckling modal ultimate bearing capacity calculation, dynamic monitoring and early warning, and early warning criteria; starting from the vibration modal change mechanism caused by the attenuation of foundation bearing capacity, the vibration mode of the tall structure is taken as the safety monitoring object, the buckling stability ultimate bearing capacity at the characteristic order corresponding to the first mutation of strain energy in the buckling modal analysis is taken as the indicator characterizing the bearing capacity of the tall structure, and the attenuation rate of the bearing capacity indicator is taken as the safety early warning indicator; the vibration and buckling modal model conversion is through the two types of models under the condition of modal consistency, using operational modal analysis to obtain the equivalent support constraint stiffness instead of the foundation constraint effect.

[0009] The required equipment includes 941B magnetoelectric low-carbon vibration pickup, sensor-specific glue, data wire, and DHDAS dynamic signal acquisition instrument.

[0010] The structural modal calibration during the construction period is as follows: obtaining the engineering data of the tall structure after completion of the construction period, and establishing a finite element geometric model of the structure-foundation-foundation of the tall structure according to the parameter information of the height, foundation form, and site strata; conducting vibration modal eigenvalue analysis of the finite element model of the structure-foundation-foundation of the tall structure, and obtaining the modal parameters of the modal vibration mode, modal curvature, and modal deformation of the first 10 natural vibration modes.

[0011] The structural status modal detection records the rod with the largest modal displacement as the inertial acceleration sensor layout point based on the modal deformation of the first 10 natural vibration modes calculated in the structural modal calibration of the construction period, and pastes the inertial acceleration sensor on the layout point. The vibration acceleration time history curve of the existing structure of the tall structure is obtained by using the environmental excitation method; the Hilbert-Huang algorithm is used to identify the modal parameters of the natural frequency, damping ratio and modal vibration mode of the tall structure.

[0012] The operational modal model is calibrated as follows: the modal confidence criterion of modal mode-modal curvature is adopted, the modal change ratio of modal curvature is used as the modal confidence parameter, and the measured modal parameters are compared with the initial modal parameters to determine the location MAC ( , );

[0013] ,

[0014] In the formula, , are the a-th and e-th order vectors of the vibration matrix, respectively. a and e are the values ​​of the a-th and e-th order calculated vibration modes at the degrees of freedom corresponding to n sensors. T is the vector transpose symbol, is the transpose of the a-th order vector of the vibration mode matrix;

[0015] The modal curvature will have a sudden change at the damage location. In addition, the modal curvature can also be used to define the curvature modal change ratio, which is more obvious than the modal curvature identification result as a derivative of the modal curvature.

[0016] ,

[0017] In the formula, represents the curvature modal change ratio, and denote the curvature modes in lossless and lossy cases, respectively.

[0018] The buckling modal analysis model is established as follows: the buckling modal analysis model replaces the constraint effect of the foundation with the equivalent support constraint. The conversion between the two is to obtain the equivalent support constraint stiffness under the condition of modal consistency through two types of models, and adopt operational modal analysis. The equivalent support constraint stiffness starts from the nearly rigid fixed end support constraint state, and gradually reduces the equivalent support constraint stiffness to obtain the equivalent support constraint stiffness matching the target modal parameters; the buckling calculation model of the existing high-rise structure adopts the stiffness parameters characterizing the local damage to correct the numerical model of the upper structure, and its lower foundation is characterized by the equivalent support constraint stiffness of the corresponding conditions.

[0019] The buckling modal ultimate bearing capacity is calculated as:

[0020] For the upper structure, the “consistent defect method” is used to correct the node coordinates, perform eigenvalue buckling analysis, read the displacement of each order of buckling mode of the structure, apply the initial defect of the structure span L / 300 for each order mode, perform nonlinear analysis to obtain the ultimate bearing capacity of the structure, calculate the actual ultimate bearing capacity of the structure through wind load logic search, and compare it with the ultimate bearing capacity obtained by the consistent defect method to determine the ultimate bearing capacity of the structure under the defect mode;

[0021] For the lower structure, combined with the measured data, the working state modal analysis is used to correct the three-way connection stiffness of the equivalent support, and the stable ultimate bearing capacity of the corrected model is calculated; according to the modal strain energy theory, the ultimate bearing capacity coefficient corresponding to the overall mutation order of the modal strain energy is found, and the allowable bearing capacity of the transmission tower is calculated.

[0022] The dynamic monitoring and early warning, i.e. the estimation of the structural bearing capacity attenuation, is carried out in a continuous cycle according to the progress of the construction of the shallow tunnel, including the current structural modal detection, the operational modal model calibration, and the establishment of the buckling modal analysis model.

[0023] The early warning criterion uses the attenuation rate of the buckling stability ultimate bearing capacity as a judgment indicator. The prediction method includes collecting tunnel construction parameters to obtain sample data, preprocessing the sample data, introducing the Bayesian optimization method to optimize the hyperparameters of the natural gradient boosting prediction model, constructing and training an optimized natural gradient boosting prediction model with the output variable as the control target based on the normalized sample data set, predicting the output variable through the optimized natural gradient boosting prediction model, outputting the optimized natural gradient boosting prediction result, and dividing the safety risk level of shallow buried tunnel construction close to tall structures according to the prediction result.

[0024] Beneficial effects: The technical solution provided by the present invention has the following beneficial effects: the influence of the attenuation of the bearing capacity of the tall structure during the construction of the shallow buried tunnel is revealed by measuring the vibration modal changes of the tall structure, and the attenuation rate of the buckling stability limit bearing capacity is used as a judgment indicator for early warning, avoiding the lag of the traditional monitoring method of the upper structure with stress, strain, deformation and displacement as reference quantities. It solves the difficulties in the implementation of underground foundation structure monitoring of existing tall structures. The traditional monitoring method of the upper structure with stress, strain, deformation and displacement as reference quantities has a safety warning lag deviation, and cannot reflect the overall bearing capacity attenuation of the tall structure. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of an embodiment of a method for identifying and early warning safety data of shallow buried tunnel construction close to tall structures of the present invention;

[0026] Figure 2 is the wind load logistic curve;

[0027] Figure 3 is the modal strain energy curve;

[0028] Figure 4 This is the result graph of the natural gradient boosting prediction model, where maximum likelihood estimation and continuous graded probability scoring are scoring rules used to calibrate the data. Figure 4 (a) and (c) are the predicted results of the true value. Figure 4 (b) and (d) are the prediction results of the predicted values. DETAILED DESCRIPTION

[0029] The safety data identification and early warning method for shallow buried tunnel construction close to tall structures of the present invention comprises eight parts, namely, safety monitoring equipment, structural modal calibration during construction, structural status modal monitoring, operational modal model calibration, buckling modal analysis model establishment, buckling modal ultimate bearing capacity calculation, dynamic monitoring and early warning, and early warning criteria; starting from the vibration modal change mechanism caused by the attenuation of foundation bearing capacity, the vibration mode of the tall structure is taken as the safety monitoring object, the buckling stability ultimate bearing capacity at the characteristic order corresponding to the first mutation of strain energy in the buckling modal analysis is taken as the indicator characterizing the bearing capacity of the tall structure, and the attenuation rate of the bearing capacity indicator is taken as the safety early warning indicator; the conversion between vibration and buckling modal models is carried out by using operational modal analysis under the condition of modal consistency of the two types of models to obtain the equivalent support constraint stiffness instead of the constraint effect of the foundation.

[0030] The construction of shallow buried tunnels will closely pass through the bearing zone of the foundation bearing capacity of tall structures, which will cause the tall structures to attenuate.

[0031] The safety monitoring equipment adopts DHDAS dynamic signal acquisition instrument, sampling frequency of 200Hz and analysis software DASPV11; 941B type magnetoelectric low-carbon vibration pickup is used to test the acceleration, velocity and displacement of the high-rise structure; the vibration pickup layout principle is based on finite element modal analysis, and is arranged at the point where the modal vibration displacement is maximum, and is encrypted around it, and the vibration pickup direction is correctly set according to the modal vibration shape and acceleration collection direction; the vibration pickup is rigidly connected to the structure, and the vibration pickup sensitivity is calibrated; other auxiliary equipment also includes communication cables, adapters, timers, metronomes, computers, power supplies and cables.

[0032] Modal calibration of structures in the completion period: obtain the engineering data of high-rise structures after completion, and establish a finite element geometric model of the structure-foundation-soil of the high-rise structure based on the parameter information; the structure of the high-rise structure adopts rod units, the foundation and subsoil adopt solid units, the soil adopts the Mohr-Coulomb constitutive model, and the concrete and steel adopt the elastoplastic DP constitutive model; carry out vibration modal eigenvalue analysis of the finite element model of the structure-foundation-soil of the high-rise structure, and obtain the modal parameters such as modal vibration mode, modal curvature, modal deformation, etc. of the first 10 natural vibration modes.

[0033] Modal monitoring of the current state of the structure: Based on the modal deformation of the first 10 natural vibration modes calculated in the modal calibration of the completed structure, the rods with larger modal displacements are recorded as the layout points of the 941B type magnetoelectric low-carbon vibration pickup. The 941B type magnetoelectric low-carbon vibration pickup is pasted on the layout points with 502 glue, and the 941B type magnetoelectric low-carbon vibration pickup and the DHDAS dynamic signal acquisition instrument are connected with wires, and the wires are pasted and fixed on the main materials of the tall structure with tape. Under the action of wind load, the vibration acceleration time history curve of the tall structure is obtained. The Hilbert-Huang algorithm is used to identify the modal parameters of high-rise structures, such as the natural frequency, damping ratio, and modal vibration shape. The general process of using the Hilbert-Huang transform to identify the modal parameters of the system under environmental excitation is as follows: First, the response data of the system is subjected to empirical mode decomposition to obtain a set of intrinsic modal function components representing the modal responses of different orders of the structure; since these components include the free vibration response of the system and the forced vibration response under random excitation, the random reduction method or natural excitation technology can be applied to the intrinsic modal function components to obtain the free decay response signal of each order of the system mode. Then, the same steps as those for identifying the modal parameters of a single-degree-of-freedom linear system under impact load are applied to process the free decay response signal, and the frequencies and damping ratios of each order of the system are obtained. Since the amplitude obtained in the actual processing process is The logarithmic curve and phase curve are not ideal straight lines, so the least squares linear fitting can be performed to obtain their slopes. Figure 1 shown.

[0034] According to the modal deformation of the first 10 natural vibration modes calculated in the modal calibration of the built-up structure, the rods with larger modal displacements are recorded as the inertial acceleration sensor layout points. The inertial acceleration sensors are pasted on the layout points using sensor-specific glue, and the vibration acceleration time history curve of the existing structure of the tall building is obtained using the environmental excitation method. The Hilbert-Huang algorithm is used to identify the modal parameters of the tall building structure, such as the natural frequency, damping ratio, and modal vibration mode.

[0035] Empirical mode decomposition extracts information from the data based on the time scale characteristics of the data to be processed, without relying on a priori basis functions, and is therefore more suitable for processing nonlinear and non-stationary signals. The intrinsic mode function obtained by empirical mode decomposition needs to meet the following two conditions: 1) In the entire time series of the intrinsic mode function, the number of zero-crossing points and the number of extreme points differ by at most one, where the extreme point refers to the maximum point or the minimum point; 2) The mean of the upper envelope formed by the maximum points and the lower envelope formed by the minimum points is zero at any time within the signal time range.

[0036] In EMD, the time scale refers to the time delay between adjacent peak points of the signal to be processed. Through EMD, the signal can be decomposed into several intrinsic mode functions with different time scales and a residual. The specific decomposition process is as follows:

[0037] (1) Find the signal to be processed All the maximum and minimum points in the signal are fitted using cubic spline curves to obtain the signal The upper and lower envelopes of

[0038] (2) Calculate the average value of the upper and lower envelopes , and subtract the mean value from the original signal to obtain :

[0039] ,

[0040] (3) Determined Does it meet the conditions of the eigenmode function components? If so, then is the first intrinsic mode function component of the original signal; if the condition is not met, As the original data, repeat steps (1)-(3) k times until the obtained Satisfy the eigenmode function conditions and record ,Will As The first intrinsic mode function component of ;

[0041] (4) From the original signal Minus , and get the remainder :

[0042] ,

[0043] Will Repeat the above steps n times as the original data until the obtained The method is further decomposed. At this point, the signal to be processed It is decomposed into a set of intrinsic mode function components and a residual, and the original signal is expressed as: ,

[0044] For a single degree of freedom linear system, the displacement response under impact load is as follows:

[0045] ,

[0046] Where: , , are the undamped circular frequency, damped circular frequency and damping ratio of the system respectively; and are the amplitude and phase corresponding to t=0 respectively, and t corresponds to time.

[0047] Signal The Hilbert transform of , then its analytical signal It is expressed as:

[0048] ,

[0049] Where: is the index of the IMF component, the amplitude and Phase The relationship is:

[0050] ,

[0051] ,

[0052] Taking the logarithm of the amplitude in the above formula and taking the derivative with respect to time, we get the following form:

[0053] ,

[0054] ,

[0055] The phase derivative of the above formula is as follows:

[0056] ,

[0057] The relationship between the undamped natural frequency, the damped natural frequency and the damping ratio is

[0058] ,

[0059] Where: , , are the i-th order undamped natural frequency, damped natural frequency and damping ratio of the system, respectively.

[0060] The undamped frequency and damping ratio can be obtained from the relationship between the three:

[0061] ,

[0062] Operational modal model calibration: characterize the damage state by the attenuation of the structural stiffness at the local damage location, gradually reduce the structural stiffness at the local damage location from the initial design conditions, and use operational modal analysis to obtain the structural stiffness parameters at the local damage location that match the actual modal parameters of the measured high-rise structure. Use the modal confidence criterion of the modal vibration mode (modal curvature), use the modal change ratio of the modal curvature as the modal confidence parameter, and repeatedly compare the measured modal parameters with the initial modal parameters to determine the location of the local structural damage of the existing transmission tower;

[0063] ,

[0064] In the formula, , are the a-th and e-th order vectors of the vibration matrix, respectively. a and e are the values ​​of the a-th and e-th order calculated vibration modes at the degrees of freedom corresponding to n sensors. T is the vector transpose symbol, is the transpose of the a-th order vector of the vibration mode matrix;

[0065] The modal curvature can have a sudden change at the damage location. In addition, the modal curvature can also be used to define the curvature modal change ratio, which is more obvious than the modal curvature identification result as a derivative of the modal curvature.

[0066] ,

[0067] In the formula, represents the curvature modal change ratio, and denote the curvature modes in lossless and lossy cases, respectively.

[0068] The Fibonacci search method is used in the optimization search process to compare the modal parameter indicators before and after correction. The optimization goals are to control the frequency frequency difference rate between the corrected modal frequency and the target modal frequency within 5%, and the modal confidence criterion between the corrected modal strain energy and the target modal strain energy within 10%.

[0069] Buckling modal analysis model establishment: First, the vibration mode needs to be converted into a buckling modal calculation model. The vibration modal analysis model is a high-rise structure-foundation integrated numerical model. The buckling modal analysis model replaces the foundation constraint with an equivalent support constraint. The conversion between the two is to obtain the equivalent support constraint stiffness under the condition of modal consistency through the two types of models. Using operational modal analysis, the equivalent support constraint stiffness starts from the near-rigid fixed end support constraint state, and gradually reduces the equivalent support constraint stiffness to obtain the equivalent support constraint stiffness matching the target modal parameters. Finally, the buckling calculation model of the high-rise structure uses the stiffness parameters that characterize local damage to modify the numerical model of the upper structure, and its lower foundation is characterized by the equivalent support constraint stiffness of the corresponding conditions; the transmission tower buckling calculation model of the initial design conditions only changes the lower foundation to the equivalent support constraint stiffness of the corresponding conditions.

[0070] Calculation of buckling modal ultimate bearing capacity: buckling modal analysis of high-rise structures under initial design conditions and actual damage conditions is carried out respectively. The buckling modal strain energy trend diagram method of different modal orders is used to find the modal characteristic order corresponding to the first mutation of the buckling modal strain energy. The buckling nonlinear analysis under this modal characteristic order is carried out to obtain the buckling stability ultimate bearing capacity. For the upper structure, the "consistent defect method" is used to correct the node coordinates, and the eigenvalue buckling analysis is carried out. The displacement of each order of the buckling mode of the structure is read. The initial defect of the structure span L / 300 is applied to each order of mode to perform nonlinear analysis to obtain the ultimate bearing capacity of the structure. The actual ultimate bearing capacity of the structure is calculated by wind load logic search, and compared with the ultimate bearing capacity obtained by the consistent defect method to determine the ultimate bearing capacity of the structure under the defect mode; for the lower structure, combined with the measured data, the working state modal analysis is used to correct the three-way connection stiffness of the equivalent support, and the stable ultimate bearing capacity of the corrected model is calculated. According to the modal strain energy theory, the ultimate bearing capacity coefficient corresponding to the overall mutation order of the modal strain energy is found, which is convenient for calculating the allowable bearing capacity of the transmission tower.

[0071] Dynamic monitoring and early warning: According to the progress of the shallow tunnel construction, the current structural modal detection, operational modal model calibration, and buckling modal analysis model establishment are carried out in a continuous cycle.

[0072] Early warning criterion: The attenuation rate of the buckling stability ultimate bearing capacity is used as the judgment index. The prediction method includes collecting tunnel construction parameters to obtain sample data, preprocessing the sample data, introducing the Bayesian optimization method to optimize the hyperparameters of the natural gradient boosting prediction model, constructing and training an optimized natural gradient boosting prediction model with the output variable as the control target according to the normalized sample data set, predicting the output variable through the optimized natural gradient boosting prediction model, outputting the optimized natural gradient boosting prediction result, and dividing the safety risk level of shallow buried tunnel construction close to tall structures according to the prediction result.

[0073] Embodiment, this embodiment takes the transmission tower model experiment as an example, the transmission tower prototype is a 220KV "stem" type transmission tower with a section tunnel passing through, the height is 36m (including 7.48m tower legs and 28.52m tower body), the total height is 46m, the root opening is 10.4m, the lower cross arm is 12.5m, and the upper cross arm is 10.2m. The buried depth of the section tunnel varies from 15 to 45m, and the transmission towers are all large slab foundations with a foundation buried depth of 2 to 4m. The basic conditions of the towers are shown in Table 1:

[0074] ,

[0075] like Figure 1 As shown, this embodiment provides a safety monitoring and early warning method for shallow buried tunnel construction close to tall structures, including:

[0076] S1. A structure-foundation integrated finite element model is established based on the completion data of the power tower, and the eigenvalue modal analysis is performed to obtain the theoretical vibration modal parameters of the existing tower. The initial defect correction is applied through the consistent defect method to obtain the theoretical initial modal parameters of the tower, and the buckling modal analysis and ultimate bearing capacity analysis of the model with initial defects are performed to obtain the theoretical buckling modal strain energy.

[0077] This embodiment is based on the design data of 26# tower and establishes a theoretical numerical model. The consistent defect method applies the initial defect with reference to the defect amplitude of L / 300 in the "Technical Code for Space Grid Structure".

[0078] It should be noted that the detailed structural parameters of transmission towers should refer to the Technical Regulations for the Structural Design of Transmission Line Towers. The cross-sections of the main materials, diagonal materials and auxiliary materials and the shear and torsion coefficients of the connections between them, the stiffness of the end connections and other parameters are particularly important for the model.

[0079] According to the model calculation results, the theoretical ultimate wind load value of the transmission tower is 49m / s, and its thirteenth-order mode is the buckling mode. The ultimate wind pressure of the transmission tower can be calculated by the wind pressure formula. , where: is the basic wind pressure; is the ultimate wind load value.

[0080] Structural ultimate wind load , where: is the standard value of wind load; For height Wind vibration coefficient at ; is the wind load shape coefficient; is the wind pressure height variation coefficient; The overall strain energy of the buckling mode structure is output, and it can be seen that the 15th and 42nd orders are energy mutation orders, the 15th order is local instability, and the 42nd order is overall instability.

[0081] S2. A 1m×1m×1m model box is used to simulate a shallow buried tunnel. The transmission tower model is subjected to a modal test using environmental excitation. The acceleration response of multiple measuring points is tested using an accelerometer. The required equipment includes: 941B type magnetoelectric low-carbon vibration pickup, sensor-specific glue, data wire, DHDAS dynamic signal acquisition instrument, and modal test analysis software DASPV11. Considering the height of a typical transmission tower, the size of the test site, and other requirements, the geometric scale ratio of the model is finally selected as 1:36, and then the coefficient formula is substituted to calculate, and the similar parameters of the model are obtained, as shown in Table 2:

[0082] Table 2 Similarity coefficients of simulations (n=36)

[0083] ,

[0084] S3. Under environmental excitation, vibration modal data are collected, and false modes are screened out by the Hilbert-Huang method to obtain the vibration modal parameters of the actual structure, as shown in Table 3. The local damage location of the structure is determined based on the modal confidence criterion.

[0085] Table 3 Test modal frequencies of the dry-shaped transmission tower

[0086] ,

[0087] Through the vibration modal parameters of the structure, combined with the modified model, the operational modal analysis is carried out to determine the three stiffnesses and torsion parameters of the equivalent support, and the conversion of the equivalent support model is completed. The equivalent support buckling modal analysis and ultimate bearing capacity analysis are carried out to obtain the buckling mode and ultimate bearing capacity of the modified equivalent support model, and the modal strain energy is extracted.

[0088] S4. Calculate the ultimate bearing capacity of the actual transmission tower and the theoretical modified model through modal strain energy bridging. According to the construction progress of the shallow buried tunnel approach, continuous and cyclic monitoring is carried out; the attenuation rate of the buckling stability ultimate bearing capacity is used as the judgment index, the tunnel construction parameters are collected to obtain sample data, and the sample data is pre-processed, and the Bayesian optimization method is introduced to optimize the hyperparameters of the natural gradient boosting prediction model. According to the normalized sample data set, an optimized natural gradient boosting prediction model with the output variable as the control target is constructed and trained, such as Figure 4 As shown, the output variable is predicted by the optimized natural gradient boosting prediction model, and the optimized natural gradient boosting prediction result is output, and the safety risk level of shallow buried tunnel construction close to tall structures is divided according to the prediction result.

[0089] The above embodiments are only used to introduce the technical solutions of the present application in detail, but the description of the above embodiments is only used to help understand the methods of the embodiments of the present invention and should not be understood as limiting the embodiments of the present invention. Changes or substitutions that can be easily thought of by technicians in this technical field should be included in the protection scope of the embodiments of the present invention.

Claims

1. A safety data identification and early warning method for shallow buried tunnel construction close to tall structures, characterized by: The safety data identification and early warning method includes eight parts, namely, safety monitoring equipment, structural modal calibration during construction, structural status modal monitoring, operational modal model calibration, buckling modal analysis model establishment, buckling modal ultimate bearing capacity calculation, dynamic monitoring and early warning, and early warning criteria; Starting from the vibration mode change mechanism caused by the attenuation of foundation bearing capacity, the vibration mode of high-rise structures is taken as the safety monitoring object, the buckling stability limit bearing capacity at the characteristic order corresponding to the first mutation of strain energy in buckling mode analysis is used as the index to characterize the bearing capacity of high-rise structures, and the attenuation rate of the bearing capacity index is used as the safety warning index; the vibration and buckling mode model conversion is to use the two types of models under the condition of modal consistency, and use the operational modal analysis to obtain the equivalent support constraint stiffness to replace the foundation constraint effect; The structural modal calibration of the construction period is as follows: obtaining the engineering data of the high-rise structure after the completion of the construction period, and establishing a finite element geometric model of the structure-foundation-foundation of the high-rise structure according to the parameter information of the height, foundation form, and site stratum; carrying out vibration modal eigenvalue analysis of the finite element model of the structure-foundation-foundation of the high-rise structure, and obtaining the modal parameters of the modal vibration mode, modal curvature, and modal deformation of the first 10 natural vibration modes; The structural status modal monitoring records the rod with the largest modal displacement as the inertial acceleration sensor arrangement point based on the modal deformation of the first 10 natural vibration modes calculated in the structural modal calibration during the construction period, and pastes the inertial acceleration sensor on the arrangement point. The vibration acceleration time history curve of the existing structure of the high-rise structure is obtained by using the environmental excitation method; the Hilbert-Huang algorithm is used to identify the modal parameters of the natural frequency, damping ratio and modal vibration mode of the high-rise structure; The operational modal model is calibrated as follows: the modal confidence criterion of modal mode-modal curvature is adopted, the modal change ratio of modal curvature is used as the modal confidence parameter, and the measured modal parameters are compared with the initial modal parameters to determine the location MAC (ψ a , ψ e ); In the formula, ψ a , e are the a-th and e-th order vectors of the vibration mode matrix, respectively, a and e are the values ​​of the a-th and e-th order calculated vibration modes corresponding to the degrees of freedom of n sensors; T is the vector transpose symbol, is the transpose of the a-th order vector of the vibration mode matrix; The modal curvature will have a sudden change at the damage location. In addition, the modal curvature can also be used to define the curvature modal change ratio, which is more obvious than the modal curvature identification result as a derivative of the modal curvature. In the formula, α j represents the curvature modal change ratio, and denote the curvature modes in lossless and lossy cases respectively; The buckling modal analysis model is established as follows: the buckling modal analysis model replaces the constraint effect of the foundation with the equivalent support constraint. The conversion between the two is to obtain the equivalent support constraint stiffness under the condition of modal consistency through two types of models, and adopt the operational modal analysis. The equivalent support constraint stiffness starts from the nearly rigid fixed end support constraint state, and gradually reduces the equivalent support constraint stiffness to obtain the equivalent support constraint stiffness matching the target modal parameters; the buckling calculation model of the existing high-rise structure adopts the stiffness parameters representing the local damage to correct the numerical model of the upper structure, and its lower foundation is characterized by the equivalent support constraint stiffness of the corresponding conditions; The buckling modal ultimate bearing capacity is calculated as: For the upper structure, the "consistent defect method" is used to correct the node coordinates, perform eigenvalue buckling analysis, read the displacement of each order of buckling mode of the structure, apply the initial defect of the structure span L / 300 for each order mode, perform nonlinear analysis to obtain the ultimate bearing capacity of the structure, calculate the actual ultimate bearing capacity of the structure through wind load logic search, and compare it with the ultimate bearing capacity obtained by the consistent defect method to determine the ultimate bearing capacity of the structure under the defect mode; For the lower structure, combined with the measured data, the working state modal analysis is used to correct the three-way connection stiffness of the equivalent support, and the stable ultimate bearing capacity of the corrected model is calculated; according to the modal strain energy theory, the ultimate bearing capacity coefficient corresponding to the overall mutation order of the modal strain energy is found, and the allowable bearing capacity of the transmission tower is calculated; The dynamic monitoring and early warning, i.e. the estimation of structural bearing capacity attenuation, is carried out in a continuous cycle according to the construction progress of the shallow tunnel approach, including the current structural modal detection, operational modal model calibration, and buckling modal analysis model establishment; The early warning criterion uses the attenuation rate of the buckling stability ultimate bearing capacity as a judgment indicator. The prediction method includes collecting tunnel construction parameters to obtain sample data, preprocessing the sample data, introducing the Bayesian optimization method to optimize the hyperparameters of the natural gradient boosting prediction model, constructing and training an optimized natural gradient boosting prediction model with the output variable as the control target based on the normalized sample data set, predicting the output variable through the optimized natural gradient boosting prediction model, outputting the optimized natural gradient boosting prediction result, and dividing the safety risk level of shallow buried tunnel construction close to tall structures according to the prediction result.

Images