Deformation prediction method for supporting structure of ten-thousand-square-meter muddy deep foundation pit

Abstract

The invention discloses a ten-thousand-square-meter muddy deep foundation pit supporting structure deformation prediction method which comprises the following steps: establishing a soil-supporting interaction rheological model, simulating stress relaxation and creep effects of muddy soil in combination with a Burgers rheological model, and realizing dynamic evolution prediction of a supporting structure deformation field and an internal force field. According to the method, the particle filter algorithm is adopted to carry out denoising processing on real-time monitoring data, model parameters are updated through the Bayesian method, and the prediction precision is improved. Meanwhile, nonlinear mapping of the deformation rate and the internal force increment is established by using a random forest algorithm, and instability risk probability distribution is generated. According to the method, multi-stage collapse early warning thresholds are set, monitoring data and prediction results are fused by adopting a weighted average method, and abnormal points of a high-risk area are identified. For a high-risk area, a genetic algorithm is applied to optimize the rigidity and prestress distribution of a supporting system, effective control over deformation and internal force of a supporting structure is achieved by dynamically adjusting construction parameters, and construction safety of the deep foundation pit is ensured.

Description

Technical Field

[0001] The present invention relates to the field of deep foundation pit support, in particular to a method for predicting deformation of a 10,000 square meter-level muddy deep foundation pit support structure. Background Art

[0002] In the field of deep foundation pit support construction with an area of 10,000 square meters and rheological muddy soil on coastal mudflats, the prediction of the deformation and stress state of the support structure is crucial to ensure the safety of the project.

[0003] Rheological silt soil has the characteristics of high compressibility, strong rheology and low strength, which makes the foundation pit support structure very easy to deform, become unstable or even collapse during construction, seriously threatening the safety of the project and the stability of the surrounding environment.

[0004] Therefore, accurately predicting the deformation and internal force distribution of the support structure and optimizing the construction plan based on the prediction results have become the core research topics in this field.

[0005] Existing support structure prediction and construction methods mainly rely on traditional finite element analysis and empirical formulas, but these methods have significant limitations when dealing with the complex mechanical behavior of rheological silty soils.

[0006] Traditional models often ignore the rheological properties of soil and are unable to accurately capture the dynamic changes of soil-support interaction during construction.

[0007] In addition, the existing methods lack effective integration of real-time monitoring data, resulting in large deviations between the prediction results and the actual construction status, delayed adjustments to the construction plan, and increased safety risks.

[0008] Specifically, the strong rheological properties and high sensitivity of rheological silty soils make the prediction of deformation and stress state of support structures face three core challenges: first, how to accurately simulate the dynamic evolution process of soil-support interaction; second, how to correct the prediction model through real-time monitoring data to improve accuracy; and third, how to establish an effective collapse warning mechanism based on the prediction and monitoring results.

[0009] These challenges directly lead to the inability to timely identify the potential instability risks of the support structure during the construction process, and it is difficult to dynamically optimize construction parameters, which in turn affects the safety and efficiency of the project.

[0010] Therefore, how to establish a dynamic rheological model of soil-support interaction, integrate real-time monitoring data to correct prediction results, and combine a multi-level early warning mechanism to achieve precise navigation of the construction process has become a key issue to ensure project safety and efficiency in the construction of deep foundation pit support covering 10,000 square meters in rheological silt. Summary of the Invention

[0011] The purpose of the present invention is to solve the above problems and provide a deformation prediction method for support structures of deep foundation pits with silt of 10,000 square meters. By establishing a rheological model of soil-support interaction, integrating real-time monitoring data to correct the prediction results, and combining a multi-level early warning mechanism, accurate navigation of the construction process can be achieved.

[0012] The technical solution adopted by the present invention to solve the technical problem is:

[0013] The deformation prediction method of the support structure of a 10,000 square meter deep foundation pit with muddy soil includes the following steps:

[0014] S101 obtains real-time monitoring data of rheological soil from the construction site, combines it with preset rheological model parameters, and constructs a rheological model of soil-support interaction to obtain the initial deformation field and internal force field;

[0015] S102 updates the soil-support interaction rheological model through real-time monitoring data to generate an optimized deformation field and internal force field;

[0016] S103: if the optimized deformation field or internal force field exceeds a preset threshold, a warning signal is generated;

[0017] S104: adjusting the soil-support interaction rheological model using an optimization algorithm according to the early warning signal to obtain an optimized construction parameter set;

[0018] S105 updates the soil-support interaction rheological model based on the optimized construction parameter set, and predicts the deformation field and internal force field of the next construction stage.

[0019] Furthermore, step S101 includes:

[0020] Acquire stress, displacement, and pore water pressure data of rheological soils from construction site sensors;

[0021] Combined with the Burgers model parameters determined in the laboratory, including the viscoelastic modulus and viscosity coefficient, the initial soil-support interaction rheological model was constructed using the finite element analysis method, and the initial deformation field and internal force field of the support structure were calculated.

[0022] Furthermore, step S102 includes:

[0023] The time-stepping method is used to determine the time step according to the daily progress of the construction phase to simulate the stress relaxation and creep effects caused by soil rheology;

[0024] Update the soil stiffness matrix and damping matrix, and calculate the dynamically evolving deformation field and internal force field through finite element analysis;

[0025] The real-time monitoring data and prediction results are fused through the Kalman filter algorithm to obtain the optimized deformation field and internal force field.

[0026] Furthermore, step S103 includes:

[0027] The displacement sequence and internal force sequence of the support structure are extracted from the real-time monitoring data, and the particle filter algorithm is used to perform denoising to obtain the denoised displacement sequence and internal force sequence;

[0028] Calculate the deviation between the denoised sequence and the predicted deformation field and internal force field. If the deviation exceeds a preset threshold, generate a warning signal.

[0029] The rheological model parameters are updated based on the normal prior distribution using the Bayesian method to generate a revised soil-support interaction rheological model.

[0030] Furthermore, step S104 includes:

[0031] Obtain abnormal deformation points and internal force points in high-risk areas from early warning signals and analyze soil-support interaction characteristics;

[0032] A genetic algorithm is used to optimize the stiffness and prestress distribution of the support system with minimization of deformation and internal force as the objective function and construction cost and time as constraints to generate an optimized set of construction parameters.

[0033] Furthermore, step S105 includes:

[0034] The soil-support interaction rheological model is updated based on the optimized construction parameter set, and the dynamic evolution of the construction process is simulated using the finite element method to obtain the optimized deformation field and internal force field.

[0035] The deformation rate and internal force change trend are extracted from the optimized deformation field and internal force field. If they exceed the preset threshold, the model is corrected through real-time monitoring data to generate a dynamically adjusted set of construction control parameters.

[0036] The beneficial effects of the present invention are:

[0037] 1. The present invention establishes a rheological model of soil-support interaction and combines it with the Burgers rheological model to simulate the stress relaxation and creep effects of silty soil, thereby realizing the dynamic evolution prediction of the deformation field and internal force field of the support structure. The present invention uses a particle filter algorithm to denoise the real-time monitoring data, and updates the model parameters through the Bayesian method to improve the prediction accuracy. At the same time, the random forest algorithm is used to establish a nonlinear mapping of deformation rate and internal force increment to generate a probability distribution of instability risk. The present invention sets a multi-level collapse warning threshold, uses a weighted average method to fuse monitoring data and prediction results, and identifies abnormal points in high-risk areas. For high-risk areas, the present invention uses a genetic algorithm to optimize the stiffness and prestress distribution of the support system, and achieves effective control of the deformation and internal force of the support structure by dynamically adjusting the construction parameters to ensure the safety of deep foundation pit construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a structural flow chart of the present invention;

[0039] Figure 2 Flowchart for calculating the preliminary deformation field U0 and internal force field F0 of the present invention;

[0040] Figure 3 A schematic diagram of the soil-support interaction rheological model M1 is generated for the present invention;

[0041] Figure 4 Schematic diagram of the risk early warning system of the present invention;

[0042] Figure 5 A flow chart is generated for the construction parameter set of the present invention. DETAILED DESCRIPTION

[0043] like Figure 1 As shown in FIG, the deformation prediction method of the support structure of a 10,000-square-meter deep foundation pit with muddy soil includes the following steps:

[0044] S101 obtains real-time monitoring data of rheological soil from the construction site, combines it with preset rheological model parameters, and constructs a rheological model of soil-support interaction to obtain the initial deformation field and internal force field;

[0045] S102 updates the soil-support interaction rheological model through real-time monitoring data to generate an optimized deformation field and internal force field;

[0046] S103: if the optimized deformation field or internal force field exceeds a preset threshold, a warning signal is generated;

[0047] S104: adjusting the soil-support interaction rheological model using an optimization algorithm according to the early warning signal to obtain an optimized construction parameter set;

[0048] S105 updates the soil-support interaction rheological model based on the optimized construction parameter set, and predicts the deformation field and internal force field of the next construction stage.

[0049] like Figure 2 As shown, step S101 includes: collecting real-time data of stress, displacement and pore water pressure of rheological muddy soil from sensors at the construction site, combining the Burgers rheological model parameters determined by laboratory rheological tests, including viscoelastic moduli E1, E2 and viscosity coefficients η1, η2, to establish an initial soil-support interaction rheological model, and calculate the preliminary deformation field U0 and internal force field F0 of the support structure.

[0050] Real-time data on stress, displacement, and pore water pressure of the rheological sludge is acquired from sensors at the construction site. Combined with laboratory-determined Burgers model parameters, including the viscoelastic moduli E1 and E2 and the viscosity coefficients η1 and η2, an initial soil-support interaction rheological model is constructed, yielding a preliminary deformation field U0 and internal force field F0. Finite element analysis is used to modify the soil-support interaction rheological model based on these preliminary deformation fields U0 and F0, combined with the real-time data, to obtain updated deformation fields U1 and F1. Using a Kalman filter algorithm, the real-time data is fused with the updated deformation fields U1 and F1 to yield optimized deformation fields U2 and F2. If the displacement value of the optimized deformation field U2 exceeds a preset threshold T1, or the stress value of the internal force field F2 exceeds a preset threshold T2, a level 1 warning signal S1 is generated. Based on the first-level warning signal S1, a Bayesian network algorithm is used to analyze the spatiotemporal trends of the deformation field U2 and the internal force field F2 to obtain a collapse risk assessment result R1. Based on this collapse risk assessment result R1 and combined with the real-time data, a numerical optimization method is used to adjust the boundary conditions of the finite element model to obtain an optimized construction plan P1.

[0051] The pore water pressure sensor uses a pressure-type pore water pressure gauge: based on the principle of hydrostatic pressure, it uses a water pressure-sensitive integrated component to measure pore water pressure. When the sensor is fixed at a point underwater, the pressure of the water column above the measuring point acts on the water pressure-sensitive integrated component, causing the component's resistance to change, resulting in a voltage change, and indirectly measuring the pore water pressure at that point.

[0052] On-site sensors collect real-time stress, displacement, and pore water pressure data from the rheological sludge, providing dynamic input for the soil-support interaction model. Due to its viscoelastic properties, the rheological sludge requires a laboratory-determined Burgers model parameter description. The Burgers model combines the elastic moduli E1 and E2 and the viscosity coefficients η1 and η2 to characterize the deformation behavior of the sludge under short-term and long-term loading.

[0053] The constitutive relation of the Burgers model can be expressed as:

[0054] Stress relationship:

[0055] σ(t)=E1ε1(t)+E2ε2(t)

[0056] where ε1(t) is the strain through the Maxwell element and ε2(t) is the strain through the Kelvin element.

[0057] Strain relationship:

[0058] ε(t)=ε1(t)+ε2(t)

[0059] Constitutive relation of Maxwell element:

[0060] σ1(t)=E1ε1(t)

[0061]

[0062] Constitutive relation of Kelvin element:

[0063]

[0064] Physical meaning of the parameters: E1 (Elastic Modulus 1): Indicates the stiffness of the elastic element in the Maxwell unit. The larger the E1, the smaller the elastic deformation of the material under short-term loading. η1 (Viscosity Coefficient 1): Indicates the viscosity of the viscous element in the Maxwell unit. The larger the η1, the smaller the creep rate of the material. E2 (Elastic Modulus 2): Indicates the stiffness of the elastic element in the Kelvin unit. The larger the E2, the stronger the elastic recovery ability of the material under long-term loading. η2 (Viscosity Coefficient 2): Indicates the viscosity of the viscous element in the Kelvin unit. The larger the η2, the smaller the relaxation rate of the material.

[0065] Based on S1, a Bayesian network was used to analyze the spatiotemporal variations of U2 and F2. Combined with historical data, the Bayesian network predicted that within 24 hours, the pit wall stress would increase to 200 kPa, with a collapse probability R1 of 30%. This result guided construction adjustments, such as reducing the excavation rate and minimizing soil disturbance.

[0066] The numerical optimization method adjusts the boundary conditions of the finite element model and generates the optimized construction plan P1.

[0067] This approach leverages multi-step collaboration to ensure dynamic model updates and accurate risk assessment. Each technical component supports each other, reducing construction risks and improving project efficiency and safety.

[0068] Step S102 includes: obtaining the initial deformation field U0 and internal force field F0, establishing a soil-support interaction rheological model through finite element analysis, obtaining the support book using the time stepping method, determining the time step Δt according to the daily progress of the construction stage, simulating the stress relaxation and creep effects caused by soil rheology, updating the stiffness matrix K and the damping matrix C, and obtaining the dynamically evolving deformation field U1 and internal force field F1.

[0069]

[0070] Δt represents the time step, L represents the length of the construction section, and v represents the daily construction progress rate.

[0071] K(t)=K0e -αt

[0072] K(t) represents the time-varying stiffness matrix, K0 represents the initial stiffness matrix, α represents the relaxation coefficient, and t represents the time variable.

[0073]

[0074] C(t) represents the damping matrix, cij(t) represents the damping coefficient of the i-th degree of freedom to the j-th degree of freedom, and these coefficients may change with time t.

[0075]

[0076] U1 represents the updated deformation field, U0 represents the initial deformation field, v(t) represents the deformation rate, and t represents the time variable.

[0077]

[0078] F1 represents the updated internal force field, K(t) represents the current stiffness matrix, C(t) represents the current damping matrix, U1 represents the current displacement, and point U1 represents the current velocity.

[0079] The stiffness matrix K reflects the decay of the elastic modulus of the soil over time, and the damping matrix C represents the viscous dissipation.

[0080] The deformation and internal force data of the support structure are obtained through real-time monitoring. If the deviation between the monitoring data and the predicted results exceeds a preset threshold, the parameters of the soil-support interaction rheological model are corrected by the least squares method to obtain the optimized model parameters. Based on the optimized model parameters, the deformation field U2 and internal force field F2 of the support structure are recalculated, and the stress state of the support structure in the current construction stage is determined through finite element analysis. If the deformation field U2 or the internal force field F2 exceeds the preset multi-level warning threshold, the time step Δt is adjusted, and the soil rheological effect is re-simulated to obtain a new deformation field U3 and internal force field F3. By iteratively comparing the real-time monitoring data with the predicted results, the stability of the stress state of the support structure is judged, and the support design optimization scheme for the current construction stage is obtained. Based on the support design optimization scheme, the soil-support interaction rheological model is updated, the deformation field U4 and internal force field F4 of the next construction stage are calculated, and the dynamic evolution trend of the support structure is determined.

[0081] When acquiring the initial deformation field U0 and internal force field F0, stress and displacement data for the silty soil can be collected using on-site sensors. Sensors are placed at the interface between the support piles and the soil to record earth pressure and pile displacement during the initial construction phase. This yields the initial soil settlement distribution (U0) and the pile internal force distribution (F0). Data accuracy depends on sensor calibration and placement density; preferably, one measurement point is set every meter to ensure data coverage across the entire support area.

[0082] When using the time-stepping method to simulate soil rheological effects, the time step length Δt can be determined based on the construction schedule. For example, in a project with daily excavation of 0.5 meters, Δt is set to 24 hours. Combined with the Burgers model parameters, the stress relaxation caused by soil creep is simulated. After updating the stiffness matrix K and the damping matrix C, the calculated settlement at the top of the support pile, represented by U1, increases by 2 mm, and the internal force in the pile, represented by F1, increases by 10%.

[0083] The least squares method can be used to correct for discrepancies between real-time monitoring data and predicted results. For example, monitoring revealed actual pile settlement of 3 mm, while the U1 prediction was 2 mm, exceeding the threshold of 5%. Using the least squares method, the viscoelastic modulus E1 and viscosity coefficient η1 in the model were adjusted to make the predicted results more realistic. After these adjustments, recalculation revealed that U2 indicated a settlement of 2.8 mm and that F2 indicated a more uniform distribution of internal forces.

[0084] Multi-level warning thresholds can be set according to support structure design specifications. The first-level threshold is set at 5 mm settlement, and the second-level threshold is 8 mm. When U2 shows a settlement of 6 mm, a first-level warning is triggered. Δt is adjusted to 12 hours, and the simulation is repeated to obtain U3, where the settlement drops to 4.5 mm. F3 shows that the internal forces do not exceed the threshold, indicating that the structure is stable.

[0085] Shortening the time step can improve simulation accuracy and identify potential risks in a timely manner. For example, when iteratively comparing monitoring data with prediction results, stability can be verified through multiple simulations.

[0086] When updating the rheological model of soil-support interaction, boundary conditions can be adjusted based on the optimization solution. For example, after adding supports, the model boundary stiffness increased by 10%, and U4 and F4 were recalculated, predicting that the long-term settlement of the support piles would not exceed the design limit. The dynamic evolution trend showed that the structure remained stable during subsequent construction, and the optimization solution effectively improved construction safety.

[0087] like Figure 3 As shown, step S103 includes: extracting the displacement sequence D and internal force sequence S of the support structure from the real-time monitoring data, denoising the sequences D and S using a particle filter algorithm to obtain a denoised displacement sequence D1 and a denoised internal force sequence S1, calculating the deviations ΔD and ΔF between the denoised sequences D1 and S1 and the dynamically evolving deformation field U1 and the internal force field F1; if the deviation ΔD or ΔF exceeds a preset threshold of 5%, updating the Burgers model parameters E1, E2, η1, and η2 based on the normal prior distribution using a Bayesian method to generate a revised soil-support interaction rheological model M1.

[0088] The displacement sequence D and internal force sequence S of the support structure are acquired from the real-time monitoring system and stored as time series data to obtain an initial data set. A particle filter algorithm is used to denoise the displacement sequence D and the internal force sequence S, generating a denoised displacement sequence D1 and a denoised internal force sequence S1, thereby obtaining a denoised data set. Using finite element analysis, the dynamically evolving deformation field U1 and internal force field F1 are calculated based on the boundary conditions of the current construction phase to obtain predicted field data.

[0089]

[0090] U1 represents the deformation field vector, n represents the number of finite element nodes, α_i represents the weight coefficient of the i-th node, K_i represents the stiffness matrix, F_i represents the node force vector, β represents the damping coefficient, Represents the Laplacian of the displacement.

[0091]

[0092] F1 represents the internal force field vector, Ω represents the computational domain, B represents the strain displacement matrix, D represents the elastic matrix, and γ represents the time correlation coefficient. represents the partial derivative of displacement with respect to time.

[0093] The deviation ΔD between the denoised displacement sequence D1 and the deformation field U1, as well as the deviation ΔF between the denoised internal force sequence S1 and the internal force field F1, are calculated. If the deviation ΔD or the deviation ΔF exceeds a preset threshold, an abnormal state is marked, and deviation state data is obtained. For the abnormal state data, the Burgers model parameters, including the elastic modulus E1, elastic modulus E2, viscosity coefficient η1, and viscosity coefficient η2, are updated using a Bayesian method based on a normal prior distribution to generate a revised soil-support interaction rheological model M1, and updated model parameters are obtained. Based on the revised soil-support interaction rheological model M1 and the real-time monitoring data, the deformation field U2 and internal force field F2 for the next construction phase are predicted to obtain predicted distribution data. If the extreme value of the deformation field U2 or the internal force field F2 in the predicted distribution data exceeds a preset multi-level threshold, a warning signal is generated, and warning state data is obtained.

[0094] Acquiring the displacement series D and internal force series S of the retaining structure in a real-time monitoring system is fundamental to analyzing soil-support interaction. During deep foundation pit construction, monitoring equipment such as inclinometers and strain gauges record the displacements and internal forces of the retaining piles hourly, generating time series data. The initial data set may contain noise, such as equipment errors or environmental vibration interference. To ensure data reliability, a particle filter algorithm is used to denoise the displacement series D and internal force series S.

[0095] Particle filtering simulates state distributions using a large number of particles, eliminating abnormal fluctuations and generating denoised displacement sequences D1 and internal force sequences S1. For example, suppose the displacement sequence D at a foundation pit monitoring point contains an abnormal peak of 100 mm. After denoising, D1 stabilizes within 10 mm, reflecting the true deformation trend. This denoised data set provides high-quality input for subsequent analysis. Finite element analysis is used to calculate the deformation field U1 and internal force field F1 based on the boundary conditions during the construction phase.

[0096] Construction boundary conditions include soil deadweight, groundwater level, and construction loads. The finite element model is meshed for the retaining structure and soil. Predictions are made for U1, which represents a 5mm displacement at the top of the retaining pile, and F1, which represents an internal force of 500 kN. The predicted field data is compared with the denoised data to calculate the deviations ΔD and ΔF.

[0097] If D1 is 4 mm and U1 is 5 mm, ΔD is 1 mm; if S1 is 480 kN and F1 is 500 kN, ΔF is 20 kN. If the preset thresholds are 2 mm and 30 kN, respectively, ΔD is normal but ΔF is abnormal, marking it as an abnormal state. Deviation data guides model modification. Bayesian methods are used to update the Burgers model parameters based on abnormal data.

[0098] The Burgers model describes the rheological properties of soil, including the elastic moduli E1, E2 and the viscosity coefficients η1, η2.

[0099] Initially, E1 was 50 MPa and η1 was 1000 MPa·s. Under abnormal conditions, E1 was adjusted to 48 MPa and η1 to 950 MPa·s using a normal prior distribution. The revised model M1 more accurately reflects the soil stiffness and damping characteristics. Updating model parameters improved prediction accuracy. Based on model M1 and monitoring data, U2 and F2 were predicted for the next phase.

[0100] For example, if U2 predicts a displacement increase of 6mm at the top of a support pile and F2 of 520kN, the predicted distribution data will be used to determine whether an early warning is triggered. If the multi-level thresholds are set at 7mm displacement or 550kN internal force, and U2 and F2 are within the current limits, no early warning is required. However, if F2 increases to 560kN in the future, an early warning signal will be generated. This early warning status data supports timely adjustments to the construction plan.

[0101] This method forms a closed loop through monitoring, denoising, modeling, deviation analysis, parameter correction, and prediction. Each link supports the others, ensuring accurate support structure condition assessment and timely dynamic adjustments. Iterative optimization of real-time monitoring and prediction provides data assurance for construction safety.

[0102] like Figure 4As shown, step S103 includes: calculating the deformation rate V and internal force increment ΔF1 of the support structure at each construction stage from the modified soil-support interaction rheological model M1, using the random forest algorithm to train the historical monitoring data D, S and the prediction results V, ΔF1, establishing a nonlinear mapping function f(V, ΔF1) of the deformation rate V and the internal force increment ΔF1, and generating the support structure instability risk probability distribution P.

[0103] The deformation rate V and internal force increment ΔF1 of the support structure at each construction stage are obtained from the soil-support interaction rheological model M1. Finite element analysis is used to calculate the deformation and internal force distribution of the support structure, obtaining initial prediction results. The initial prediction results are revised based on real-time monitoring data, and the soil-support interaction rheological model M1 is updated using an information-based construction method to obtain the revised deformation rate V and internal force increment ΔF1.

[0104]

[0105] V(t) represents the deformation rate of the support structure at time t, u(t) represents the displacement function, α is the correction coefficient, d_i represents the displacement increment of the i-th monitoring point, and t_i represents the corresponding time interval.

[0106]

[0107] ΔF1 represents the internal force increment of the support structure, Ks represents the structural stiffness matrix, Δε represents the strain increment, β is the stress correction coefficient, Pj represents the load value of the jth measuring point, and Aj represents the corresponding action area.

[0108] The random forest algorithm is used to train the historical monitoring data D, S and the corrected deformation rate V and internal force increment ΔF1, and a nonlinear mapping function f(V, ΔF1) is established, where V represents the deformation rate and ΔF1 represents the internal force increment, and the mapping relationship is obtained.

[0109]

[0110] f(V, ΔF1) represents the nonlinear mapping function, w_k represents the weight of the kth decision tree in the random forest, h_k represents the output of the kth decision tree, and T represents the total number of decision trees. Based on the nonlinear mapping function f(V, ΔF1), the support structure instability risk probability distribution P is calculated to obtain the risk probability value.

[0111]

[0112] P represents the probability of instability risk, M represents the number of sampling times, f_j represents the j-th prediction result, θ represents the instability threshold, and I represents the indicative function.

[0113] If the risk probability exceeds the preset threshold T1, an information-based construction method is used to adjust the support design parameters, update the soil-support interaction rheological model M1, and recalculate the deformation rate V and internal force increment ΔF1 to obtain an optimized prediction result. Based on the optimized prediction results and the multi-level warning threshold, the support structure instability risk level is determined and a corresponding warning signal is generated. The accuracy of the warning signal is verified based on real-time monitoring data, and the instability risk probability distribution P is updated using finite element analysis to obtain the final risk assessment result.

[0114] When obtaining the support structure's deformation rate V and internal force increment ΔF1 in the soil-support interaction rheological model M1, the support structure's displacement and internal force data can be collected through a real-time monitoring system. For example, in a foundation pit project, the support structure is steel sheet piles. Monitoring points record displacement data hourly, generating a displacement sequence D. The deformation rate V is calculated through time difference calculation. For example, if the displacement change at a point in 24 hours is 12 mm, then V is 0.5 mm / hour. The internal force increment ΔF1 is converted to an internal force increment by measuring the stress change in the steel sheet pile using strain gauges. For example, if the internal force in a section of steel sheet pile increases from 1000 kN to 1200 kN, ΔF1 is 200 kN. This data collection method ensures the accuracy of the model input.

[0115] When using finite element analysis to calculate the deformation and internal force distribution of the support structure, a two-dimensional or three-dimensional numerical model can be created based on the M1 model. Assuming a foundation pit depth of 10 meters and clay soil, the model sets the elastic modulus of the steel sheet piles and the shear modulus of the soil. Finite element analysis generates initial predictions, such as a maximum deformation of 15 mm and an internal force of 1300 kN at a specific point.

[0116] When revising the initial prediction results based on real-time monitoring data, the information-based construction method adjusts the model parameters by comparing the monitoring data with the prediction results.

[0117] Monitoring data showed a deformation of 18 mm at a certain point, exceeding the predicted value by 15 mm. An iterative optimization algorithm was then used to update the soil stiffness parameters, generating a revised V of 0.6 mm / h and ΔF1 of 220 kN. This correction improved the model's adaptability.

[0118] Using a random forest algorithm to train historical monitoring data D, S, and the corrected V and ΔF1, a model consisting of 100 decision trees can be constructed. Assume that the historical data contains 1,000 sets of samples, each including displacement, internal force, deformation rate, and internal force increment. After training, a nonlinear mapping function f(V, ΔF1) is obtained. For example, if the input V is 0.5 mm / h and ΔF1 is 200 kN, the output instability risk probability P is 0.3. This algorithm can capture complex data relationships and improve the accuracy of risk prediction.

[0119] In one embodiment, when calculating the probability distribution of instability risk, P, if P exceeds a threshold value, T1—for example, if T1 is set to 0.4 and a calculated P value is 0.45—then a design parameter adjustment is triggered. An information-based construction approach might increase the embedded depth of the steel sheet pile from 8 meters to 10 meters. After updating the M1 model and recalculating, the result is a decrease in V to 0.4 mm / hour and P to 0.35. This adjustment reduces the risk of instability.

[0120] To determine the risk level of support structure instability, we set multiple warning thresholds based on the optimized V and ΔF1. For example, a V exceeding 0.7 mm / hour or a ΔF1 exceeding 300 kN indicates a high risk. For example, if V at a certain point is 0.8 mm / hour, a high-risk warning signal is generated. This tiered warning system facilitates construction management.

[0121] When verifying the accuracy of the early warning signal, real-time monitoring data showed that V at a certain point was 0.75 mm / hour, consistent with the warning. After updating P through finite element analysis, the risk level was confirmed to be high. The final assessment results guided reinforcement measures, such as adding support beams. This verification ensured the reliability of the risk assessment.

[0122] like Figure 4 As shown, step S103 includes: obtaining a risk probability value from the instability risk probability distribution P, setting the third-level collapse warning thresholds T1 = 80%, T2 = 60%, and T3 = 40%. If the probability value P exceeds the first-level threshold T1, the denoised displacement sequence D1, denoised internal force sequence S1, and the predicted deformation field U1 and internal force field F1 are fused by the weighted average method based on the monitoring data variance σD, σS and the prediction model confidence α to generate abnormal deformation points U2 and abnormal internal force points F2 in the high-risk area and the warning signal A.

[0123] Obtain a risk probability value from the instability risk probability distribution P, the risk probability value P represents the instability of the support structure; compare the risk probability value P with the preset three-level warning thresholds T1, T2, and T3 to determine whether it exceeds T1, and obtain the triggering warning condition, wherein T1, T2, and T3 represent thresholds of different risk levels; if the risk probability value P exceeds T1, obtain the monitoring data variance σD, σS and the prediction model confidence α, wherein σD represents the variance of the displacement monitoring data, σS represents the variance of the internal force monitoring data, and α represents the reliability of the prediction model; adopt the weighted average method to fuse the denoised displacement sequence D1, the denoised internal force sequence S1, the predicted deformation field U1, and the internal force field F1 to obtain the fused data sequence D2, S2, U2, and F2, wherein D1 and S1 represent the denoised displacement and internal force sequences, and U1 and F1 represent the predicted deformation and internal force distribution; through the soil-support interaction rheological model combined with finite element analysis, based on the fused data sequence D2, S2, U2, and F2 , calculate the deformation distribution and internal force distribution of the support structure, and obtain the predicted deformation point U3 and internal force point F3; use real-time monitoring data to perform deviation analysis on the predicted deformation point U3 and internal force point F3, calculate the deviation values ΔU and ΔF, and judge whether the deviation values ΔU and ΔF exceed the preset thresholds to obtain the corrected deformation point U4 and internal force point F4, and the ΔU and ΔF represent the deviations between the predicted and actual deformation and internal force; based on the corrected deformation point U4 and internal force point F4, extract the abnormal deformation point U5 and the abnormal internal force point F5, and generate an early warning signal A1, and the U5 and F5 represent deformation and internal force points beyond the normal range; based on the early warning signal A1 and real-time monitoring data, optimize the support design parameters and generate an adjusted support scheme B1; for the adjusted support scheme B1, recalculate the soil-support interaction rheological model, update the predicted deformation field U6 and internal force field F6, judge whether the safety threshold is met, and generate a final early warning signal A2, and the U6 and F6 represent the updated deformation and internal force distribution.

[0124] The risk probability value is obtained from the instability risk probability distribution P. For example, in a foundation pit support project, a P value of 0.75 indicates a high probability of instability in the support structure. In principle, the P value is derived through statistical analysis based on historical monitoring data and prediction models, reflecting the potential risk of the structure. Comparing the P value with the three-level warning thresholds T1 = 0.6, T2 = 0.8, and T3 = 0.9, the P value exceeds T1 but is lower than T2, triggering a level one warning, indicating that further analysis is required but no immediate work stoppage is required.

[0125] In one possible implementation, if the P value exceeds T1, the monitoring data variances σD and σS, as well as the prediction model confidence α, are obtained.

[0126] For example, the displacement monitoring data variance σD=0.02mm 2, internal force monitoring data variance σS=1.5kN 2 , the confidence level α = 0.95, indicating that the data fluctuation is small and the model is reliable. The variance reflects the degree of dispersion of the monitoring data, and the confidence level measures the stability of the prediction model, which helps to determine whether the data is reliable.

[0127] Specifically, the weighted average method is used to fuse the denoised displacement sequence D1, the denoised internal force sequence S1, the predicted deformation field U1 and the internal force field F1 to generate the fused data sequences D2, S2, U2 and F2.

[0128] For example, D1 contains the displacement sequence of the top of the foundation pit, and S1 is the internal force sequence of the anchor rod, with weights of 0.4, 0.3, 0.2, and 0.1, respectively. After fusion, smoother D2 and S2 are obtained, which reduces noise interference and improves data consistency.

[0129] By combining the soil-support interaction rheological model with finite element analysis, the deformation and internal force distribution are calculated based on D2, S2, U2, and F2, and the predicted deformation point U3 and internal force point F3 are obtained.

[0130] For example, U3 shows the maximum deformation in the middle of the foundation pit is 15mm, and F3 shows the internal force of a certain anchor is 120kN. These points reflect the stress state of the support structure and provide a basis for subsequent corrections.

[0131] For example, based on real-time monitoring data, deviation analysis of U3 and F3 is performed to calculate the deviation values ΔU and ΔF. Assuming the actual monitored deformation is 16 mm and the internal force is 125 kN, ΔU = 1 mm and ΔF = 5 kN. If the preset thresholds are ΔU ≤ 2 mm and ΔF ≤ 10 kN, the deviation is within the acceptable range. After correction, U4 = 15.5 mm and F4 = 122 kN, indicating that the prediction results are relatively accurate.

[0132] Based on U4 and F4, outlier points U5 and F5 are extracted. Assuming U5 represents a point where local deformation exceeds 20 mm, and F5 represents a point where internal force exceeds 150 kN, a Level 1 warning signal A1 is generated, indicating that the local area requires attention. A1 is triggered based on outlier points, facilitating timely identification of potential risks.

[0133] Based on A1 and real-time monitoring data, the support design parameters are optimized and the adjustment plan B1 is generated.

[0134] For example, increasing anchor density or adjusting prestressing force can generate the B1 solution. This optimization reduces risk and improves structural stability by adjusting parameters.

[0135] The soil-support interaction rheological model is recalculated for B1, and the predicted deformation field U6 and internal force field F6 are updated.

[0136] For example, the maximum deformation of U6 dropped to 12mm, and the internal force of F6 dropped to 110kN, meeting the safety threshold, generating the final warning signal A2 as "safe." This iterative update ensures that the predicted results are consistent with the actual working conditions, ensuring construction safety.

[0137] like Figure 5 As shown, step S104 includes: obtaining abnormal deformation points U2 and abnormal internal force points F2 in high-risk areas from the early warning signal A, analyzing the soil-support interaction characteristics of the corresponding area, and using a genetic algorithm to minimize deformation U2 and internal force F2 as the objective function and construction cost C and time T as constraints to optimize the support system stiffness K1 and prestress distribution P1, and generate an optimized construction parameter set Q.

[0138] Abnormal deformation point U2 and abnormal internal force point F2 are obtained from warning signal A. Finite element analysis is used to calculate the deformation distribution V1 and internal force distribution W1 of the support structure using a rheological model of soil-support interaction, resulting in the predicted deformation field Y1 and internal force field Z1. Based on the predicted deformation field Y1 and internal force field Z1, combined with real-time monitoring data M1, the deformation deviation ΔY and internal force deviation ΔZ are calculated. If the deviation ΔY or ΔZ exceeds the preset threshold T4, a weighted average method is used to fuse the real-time monitoring data M1 with the predicted deformation field Y1 and internal force field Z1 to generate the corrected deformation field Y2 and internal force field Z2. A genetic algorithm is used to optimize the support stiffness K2 and prestress distribution P2 of the support structure, with minimizing the corrected deformation field Y2 and internal force field Z2 as the objective function and construction cost C1 and construction time T1 as constraints, resulting in the optimized parameter set Q1. Finite element analysis (FEM) is used to calculate the deformation distribution V2 and internal force distribution W2 of the support structure using a rheological model of soil-support interaction based on the optimized parameter set Q1, resulting in an updated deformation field Y3 and internal force field Z3. The monitoring deformation point N1 and the monitoring internal force point N2 are obtained from real-time monitoring data M1. The deviation ΔY1 between the updated deformation field Y3 and the monitoring deformation point N1, as well as the deviation ΔZ1 between the internal force field Z3 and the monitoring internal force point N2, are calculated. If the deviation ΔY1 or ΔZ1 exceeds a preset threshold T5, the updated deformation field Y3, the internal force field Z3, and the monitoring deformation point N1 and the monitoring internal force point N2 are fused using a weighted average method to generate the corrected deformation field Y4 and internal force field Z4. Based on the corrected deformation field Y4 and internal force field Z4, abnormal deformation points U3 and abnormal internal force points F3 are extracted. If the abnormal deformation point U3 or abnormal internal force point F3 exceeds the warning threshold T6, an early warning signal A3 is generated. Through the information construction method, based on the early warning signal A3 and the corrected deformation field Y4 and internal force field Z4, the support design parameters are adjusted to generate the optimized support scheme B2.

[0139] Specifically, obtaining the abnormal deformation point U2 and abnormal internal force point F2 from the early warning signal A is the core step in the risk assessment of the support structure.

[0140] Based on real-time monitoring data and predictive models, early warning signal A identifies areas of potential instability within the support structure, such as localized points of excessive deformation in the tunnel lining. Abnormal deformation points U2 represent coordinates where displacement exceeds the normal range. For example, a monitoring point with a displacement of 5 mm exceeds the threshold of 3 mm. Abnormal internal force points F2 might indicate a stress of 200 kPa in a specific section, exceeding the safety limit of 150 kPa. These points are generated through a combination of sensor data and predictive models to accurately locate high-risk areas.

[0141] The soil-support interaction rheological model combined with finite element analysis is used to calculate the deformation distribution V1 and internal force distribution W1 of the support structure, and the predicted deformation field Y1 and internal force field Z1 are obtained.

[0142] Based on Y1 and Z1, combined with real-time monitoring data M1, the deviations ΔY and ΔZ are calculated. For example, M1 shows that the actual deformation of a point is 4.5 mm, while Y1 predicts 4 mm, and ΔY is 0.5 mm. If ΔY exceeds a threshold T4, such as 0.3 mm, a correction is triggered.

[0143] A weighted averaging method is used to fuse M1 with Y1 and Z1 to generate the corrected deformation field Y2 and the internal force field Z2. The weighted averaging method assigns weights based on data reliability. For example, a weight of 0.6 for M1 and 0.4 for Y1 results in an adjusted deformation point of 4.3 mm in Y2. This fusion improves prediction accuracy.

[0144] Genetic algorithm is used to optimize support stiffness K2 and prestress distribution P2, with Y2 and Z2 as objective functions and construction cost C1 and time T1 as constraints.

[0145] The initial K2 was 5 GPa, which was increased to 6 GPa after optimization. The prestress P2 was adjusted from 100 kPa to 120 kPa. A genetic algorithm iteratively searched for the global optimal solution, ensuring a cost-effective and efficient support solution. The optimized parameter Q1 was used in finite element analysis to generate updated deformation fields Y3 and internal force fields Z3. The deviations ΔY1 and ΔZ1 were further reduced, for example, to 0.2 mm.

[0146] If ΔY1 or ΔZ1 exceeds threshold T5, such as 0.15 mm, Y3 and Z3 are fused again with monitoring points N1 and N2 to generate Y4 and Z4. After fusion, the abnormal deformation point U3 and internal force point F3 are extracted. If U3 deformation reaches 5.5 mm, exceeding the T6 threshold of 5 mm, warning signal A3 is triggered.

[0147] Informatized construction adjusts design parameters based on A3, Y4, and Z4, such as increasing anchor density, to generate optimized solution B2. This approach ensures the safety and stability of the support structure through multiple rounds of correction and optimization.

[0148] The above process, from outlier extraction to optimization solution generation, is a progressive and interconnected process. Each step relies on real-time data and predictive models, balancing accuracy and efficiency to provide reliable support for support design in complex geological conditions.

[0149] Step 105: Update the soil-support interaction rheological model M1 from the optimized construction parameter set Q, use the finite element method to re-simulate the dynamic evolution of the construction process, calculate the optimized support structure deformation field U3 and internal force field F3, and verify the control effect of the optimized parameter set Q on the deformation field U3 and internal force field F3.

[0150] Obtain the initial construction parameter set Q and the soil-support interaction rheological model M1. Generate an updated model M2 by updating the parameters of model M1. Use the finite element method to simulate the dynamic evolution of model M2 to obtain the deformation field U3 and internal force field F3 of the support structure. Through real-time monitoring data, extract the actual deformation data D1 and internal force data F1 of the support structure to generate a monitoring data set S1. If the deviation between the monitoring data set S1 and the deformation field U3 and internal force field F3 exceeds the preset threshold, adjust the parameters of model M2 through the information construction method to generate a corrected model M3. Based on the corrected model M3, use the finite element method to simulate the construction process to obtain the optimized deformation field U4 and internal force field F4. Through the deformation field U4 and internal force field F4, combined with the preset multi-level warning threshold, judge the stress state of the support structure and generate a warning signal W1.

[0151] Specifically, obtaining the initial construction parameter set Q and the soil-support interaction rheological model M1 is the basis for optimizing support design. The initial construction parameter set Q includes support stiffness, prestress distribution, and material parameters.

[0152] For example, the support stiffness can be set to 1000 kN / m, the prestressing force to 500 kPa, and the concrete strength grade to C30. The soil-support interaction rheological model M1 describes the mechanical coupling relationship between the soil and the support structure. It is constructed based on the elastic foundation beam theory, combined with the soil elastic modulus of 20 MPa and Poisson's ratio of 0.3.

[0153] M1 is calibrated using on-site geological survey data to ensure that the initial model is close to the actual working conditions. The updated model M1 to generate M2 must incorporate real-time construction data.

[0154] The parameters of M1 are updated by monitoring soil settlement and support internal forces. For example, if monitoring reveals that the soil settlement rate exceeds expectations and reaches 5 mm / day, the soil stiffness parameters in M1 are adjusted to generate M2. The soil elastic modulus may be adjusted from 20 MPa to 18 MPa.

[0155] The updating process needs to ensure the convergence of the model and avoid the calculation instability caused by parameter adjustment. The finite element method is used to simulate the dynamic evolution of M2 to obtain the deformation field U3 and the internal force field F3.

[0156] Finite element analysis can be performed using commercial software such as ANSYS to simulate the stress evolution of the support structure during construction. Assuming the simulation results show a maximum deformation of 8 mm and a maximum internal force of 600 kN, the corresponding U3 and F3 fields are generated.

[0157] The dynamic evolution simulation considers a time step of 1 day and simulates the construction progress over 7 days to capture the spatiotemporal distribution of deformation and internal forces. The actual deformation data D1 and internal force data F1 are extracted from real-time monitoring data to generate the monitoring dataset S1.

[0158] The monitoring equipment includes displacement gauges and stress sensors, which record a deformation of 7.5mm and an internal force of 620kN at a certain point in the support structure, forming S1. The monitoring frequency can be set to twice a day to ensure the timeliness of the data.

[0159] The accuracy of S1 directly affects the reliability of subsequent deviation analysis. If the deviation between S1 and U3 and F3 exceeds a preset threshold, such as a deformation deviation threshold of 2 mm or an internal force deviation threshold of 50 kN, the M2 parameters are adjusted to generate M3.

[0160] For example, the deviation analysis shows that the deformation deviation is 2.5mm, triggering an adjustment.

[0161] By combining the parameters of S1 and M2 using a weighted average method, M3 is generated. The adjusted soil stiffness may be 17.5 MPa. M3 improves the model's fit to actual working conditions. Finite element method simulations of the construction process based on M3 yielded optimized U4 and F4.

[0162] The simulation shows that the maximum deformation of U4 is reduced to 6.5 mm, and the maximum internal force of F4 is 580 kN, indicating that the support structure is more balanced after optimization.

[0163] The simulation process can refine the grid division and improve the calculation accuracy. The stress state is judged by combining U4 and F4 with multi-level warning thresholds to generate the warning signal W1.

[0164] The warning threshold is divided into three levels: deformation less than 5mm is safe, 5-7mm is caution, and more than 7mm is dangerous. Suppose U4 shows that a certain area has a deformation of 6.8mm, which triggers the W1 signal of the caution level.

[0165] W1 can guide the construction team to locally increase support stiffness or adjust the construction schedule to reduce potential risks.

[0166] Step S105: Extract the deformation rate V1 and internal force change trend ΔF2 of the support structure from the optimized deformation field U3 and internal force field F3. If the deformation rate V1 exceeds 2 mm / day or the internal force change trend ΔF2 exceeds 10% of the design value, use the real-time monitoring data D and S to further calibrate the model M1, and cyclically execute the high-risk area anomaly point generation, construction parameter optimization, and dynamic evolution simulation to generate a dynamically adjusted construction control parameter set Q1.

[0167] The deformation rate V1 of the support structure is extracted from the deformation field U3 obtained from the finite element analysis, and the internal force variation trend ΔF2 is extracted from the internal force field F3 to obtain the initial deformation rate V1 and internal force variation trend ΔF2. If the deformation rate V1 exceeds the preset threshold or the internal force variation trend ΔF2 exceeds the design value ratio, real-time monitoring data D and S are acquired and fused using a weighted average method to obtain fused monitoring data D1. Error analysis is performed between the fused monitoring data D1 and the prediction model M1, and the parameters of the prediction model M1 are corrected using the least squares method to obtain the corrected model M2. Finite element analysis is performed based on the corrected model M2 to generate a set of outliers P1 in the high-risk area, and the distribution characteristics of the outlier set P1 are determined. Correlation analysis is performed between the outlier set P1 and the construction parameters Q1, and the construction parameters Q1 are optimized using a genetic algorithm to obtain the optimized construction parameters Q2. Dynamic evolution simulation is performed based on the optimized construction parameters Q2 to generate the prediction results R1 of the support structure deformation and internal force distribution. By comparing and analyzing the prediction result R1 with the fused monitoring data D1, the correction model M2 is updated to obtain an updated model M3.

Claims

1. A method for predicting deformation of support structures in deep foundation pits of 10,000 square meters, characterized by: The following steps are involved: S101 obtains real-time monitoring data of rheological soil from the construction site, combines it with preset rheological model parameters, and constructs a soil-structure interaction model to obtain the initial deformation field and internal force field; S102 updates the soil-structure interaction model through real-time monitoring data to generate optimized deformation field and internal force field; S103: if the optimized deformation field or internal force field exceeds a preset threshold, a warning signal is generated; S104: adjusting the soil-structure interaction model using an optimization algorithm according to the early warning signal to obtain an optimized construction parameter set; S105 updates the soil-structure interaction model based on the optimized construction parameter set, and predicts the deformation field and internal force field of the next construction stage.

2. The method for predicting deformation of a 10,000 square meter deep muddy foundation pit support structure according to claim 1, characterized in that: Step S101 includes: Acquire stress, displacement, and pore water pressure data of rheological soils from construction site sensors; Combined with the Burgers model parameters determined in the laboratory, including the viscoelastic modulus and viscosity coefficient, the initial soil-structure interaction model was constructed using the finite element analysis method, and the initial deformation field and internal force field of the support structure were calculated.

3. The method for predicting deformation of a 10,000 square meter deep muddy foundation pit support structure according to claim 1, characterized in that: Step S102 includes: The time-stepping method is used to determine the time step according to the daily progress of the construction phase to simulate the stress relaxation and creep effects caused by soil rheology; Update the soil stiffness matrix and damping matrix, and calculate the dynamically evolving deformation field and internal force field through finite element analysis; The real-time monitoring data and prediction results are fused through the Kalman filter algorithm to obtain the optimized deformation field and internal force field.

4. The method for predicting deformation of a 10,000 square meter deep muddy foundation pit support structure according to claim 1, characterized in that: Step S103 includes: The displacement sequence and internal force sequence of the support structure are extracted from the real-time monitoring data, and the particle filter algorithm is used to perform denoising to obtain the denoised displacement sequence and internal force sequence; Calculate the deviation between the denoised sequence and the predicted deformation field and internal force field. If the deviation exceeds a preset threshold, generate a warning signal. The rheological model parameters are updated based on the normal prior distribution using the Bayesian method to generate a revised soil-structure interaction model.

5. The method for predicting deformation of a 10,000 square meter deep muddy foundation pit support structure according to claim 1, characterized in that: Step S104 includes: Obtain abnormal deformation points and internal force points in high-risk areas from early warning signals and analyze soil-structure interaction characteristics; A genetic algorithm is used to optimize the stiffness and prestress distribution of the support system with minimization of deformation and internal force as the objective function and construction cost and time as constraints to generate an optimized set of construction parameters.

6. The method for predicting deformation of a 10,000 square meter deep muddy foundation pit support structure according to claim 1, characterized in that: Step S105 includes: The soil-structure interaction model is updated based on the optimized construction parameter set, and the dynamic evolution of the construction process is simulated using the finite element method to obtain the optimized deformation field and internal force field. The deformation rate and internal force change trend are extracted from the optimized deformation field and internal force field. If they exceed the preset threshold, the model is corrected through real-time monitoring data to generate a dynamically adjusted set of construction control parameters.

Images