Soft soil area vertical shaft deep foundation pit ground subsidence prediction method and device

Abstract

The invention discloses a soft soil area vertical shaft deep foundation pit ground subsidence prediction method and device, and belongs to the field of deep foundation pit engineering. The method comprises the following steps: establishing a two-dimensional finite element model for ground surface settlement analysis under the coupling action of a seepage field and a stress field; the working condition of the model is adjusted to determine main factors affecting ground surface settlement; establishing a regression equation for predicting the maximum wall deflection caused by support excavation and an empirical equation for predicting a deformation ratio based on the main factors; training and predicting parameters in the regression equation and the empirical equation by adopting a machine learning algorithm on the basis of accumulated ground surface settlement data obtained by numerical analysis of the typical construction process of the vertical shaft; determining the maximum surface settlement based on the deflection equation and the deformation ratio equation; and based on the maximum surface subsidence, according to the outside-pit subsidence influence range and the subsidence curve form, the surface subsidence influence area around the foundation pit is partitioned, and surface subsidence is predicted. According to the method and device, the ground subsidence induced by shaft foundation pit construction can be rapidly calculated.

Description

Technical Field

[0001] The present invention relates to the field of deep foundation pit engineering, and in particular to a method and device for predicting ground settlement of a vertical deep foundation pit in a soft soil area. Background Art

[0002] Pipe jacking (shield tunneling) generally requires the construction of starting and receiving working pits first, and the working pits are generally deep foundation pit projects. In foundation pit projects, with the construction of retaining structures and subsequent foundation pit dewatering and earth excavation, first, the soft clay is gradually affected by construction disturbances; second, foundation pit dewatering causes the original groundwater seepage state of the surrounding soil to change, thereby changing the water head inside and outside the foundation pit, and also causes changes in the seepage field and stress field state in the soil. The seepage field and stress field interact and influence each other; finally, after earth excavation, the soil stress is released, breaking the original stress equilibrium state, the plastic zone expands, and the pit bottom bulges, causing deformation of the foundation pit support structure and settlement of the surrounding surface. At the same time, due to the influence of pressurized water in sandy soil, the construction process is prone to phenomena such as erosion, sand flow, and sudden surges, causing engineering accidents.

[0003] When there are structures, roads, railways, underground pipelines, and other facilities around the shaft, the impact of shaft foundation pit construction on surrounding surface settlement must also be considered. Therefore, the shaft foundation pit needs to meet the deformation control requirements of the surrounding environment. This is especially true in soft soil areas with complex engineering conditions and sensitive surrounding environments. Deformation control often plays a dominant role. Foundation pit design and construction have shifted from conventional strength control to deformation control. Therefore, it is particularly important to study and summarize the deformation laws of foundation pits in soft soil areas.

[0004] Methods for calculating ground settlement induced by foundation pit construction generally include the layered summation method and numerical calculation methods. The layered summation method uses the theory of isotropic homogeneous linear deformable bodies. The effective stress change in each soil layer during foundation pit construction must be calculated before calculating the ground settlement using the layered summation method. However, this method is difficult to implement in practice. Firstly, the additional stress change in the soil caused by excavation unloading is difficult to calculate, and currently no analytical formula exists. Secondly, the groundwater level change caused by foundation pit dewatering requires numerical seepage calculations.

[0005] Numerical calculation methods require the establishment of a finite element model that includes the soil stratification of the vertical shaft foundation pit, the supporting structure, the layered excavation conditions, and surrounding structures. Appropriate soil constitutive models and calculation parameters are used to simulate the interaction between the retaining structure and the soil, thereby calculating and analyzing the impact of foundation pit construction on the surrounding environment. Finite element model assumptions, soil constitutive model selection, numerical calculation parameter values, and simulation of the interaction between the retaining structure and the soil all have a significant impact on the simulation results. The results are heavily dependent on the user's skill level, resulting in high discreteness and computational time and effort.

[0006] Therefore, it is necessary to study a prediction method and device for quickly calculating ground settlement induced by shaft foundation pit construction. Summary of the Invention

[0007] In view of the above problems, the present invention provides a method and device for predicting ground settlement of a deep foundation pit in a vertical shaft in a soft soil area.

[0008] According to one aspect of the present invention, a method for predicting ground settlement of a deep foundation pit in a soft soil area is provided, comprising the following steps:

[0009] Step S1: Based on the actual construction simulation conditions of the foundation pit, a two-dimensional finite element model of soil, support, and steel sheet piles for surface settlement analysis is established considering the coupling effect of seepage field and stress field;

[0010] Step S2: Using the control variable method to adjust the working conditions of the two-dimensional finite element model established in step S1, based on the model calculation results, analyze the factors affecting the ground settlement of the deep foundation pit in the soft soil area to determine the main factors affecting the ground settlement;

[0011] Step S3: establishing a regression equation for predicting the maximum wall deflection caused by support excavation and an empirical equation for predicting the deformation ratio based on the main factors affecting surface settlement;

[0012] Step S4: Based on the accumulated surface settlement data obtained through numerical analysis of the typical shaft construction process, a machine learning algorithm is used to train and predict the parameters in the regression equation and the empirical equation, and the deflection equation and the deformation ratio equation are determined based on the parameters;

[0013] Step S5: determining the maximum surface settlement based on the deflection equation and the deformation ratio equation;

[0014] Step S6: Based on the maximum surface settlement, the settlement influence range outside the pit and the settlement curve shape, the ground settlement influence area around the foundation pit is divided into zones and the surface settlement profile is calculated to obtain the surface settlement of any point outside the foundation pit.

[0015] According to one embodiment of the present invention, ground settlement caused by precipitation and step-by-step excavation is considered when establishing the two-dimensional finite element model in step S1.

[0016] According to one embodiment of the present invention, when establishing the two-dimensional finite element model in step S1, the initial water level is set at the side boundary of the foundation pit to simulate the hydraulic replenishment of the boundary, and the dewatering process is simulated by setting the water level boundary after the foundation pit is dewatered.

[0017] According to one embodiment of the present invention, when establishing the two-dimensional finite element model in step S1, the Geostatic module is used to apply the initial geostress field of the model.

[0018] According to one embodiment of the present invention, in step S1 , the horizontal displacement of the left and right boundaries of the model is restricted to 0, the horizontal and vertical displacements of the bottom boundary are restricted to 0, and the initial pore pressure of the upper boundary is set to 0.

[0019] According to one embodiment of the present invention, in step S1, the upper soil of the model adopts the modified Cambridge constitutive model, and the underlying rock layer adopts the Mohr-Coulomb constitutive model. The soil parameters are comprehensively determined based on the measured values ​​of indoor and outdoor tests in the survey report, the derived values ​​of empirical formulas and the numerical inversion values. The soil unit adopts the pore pressure unit, the enclosure structure adopts the CPE4 four-node bilinear plane strain quadrilateral unit, and the concrete support and steel support both adopt the beam unit B21.

[0020] According to one embodiment of the present invention, the working conditions adjusted in step S2 include excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness and soft soil layer thickness.

[0021] According to one embodiment of the present invention, step S2 increases the adjusted variables in sequence according to a fixed interval when adjusting the excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness and soft soil layer thickness, and determines whether the excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness and / or soft soil layer thickness are used as the main factors affecting the surface settlement based on the change in surface settlement calculated by the model.

[0022] According to one embodiment of the present invention, the following five input variables are used in step S3 to establish a regression equation for predicting the maximum wall deflection: H e , EI / γ w h avg 4 , B, S u / σ′ v and E i / σ′ v , where H e represents the excavation depth, EI represents the stiffness of the steel sheet pile, γ w Indicates the weight of water, h avg represents the average vertical spacing of horizontal supports, B represents the excavation width, S u represents the shear strength of soil, σ v ' represents the vertical effective stress, E i represents the initial Young's modulus.

[0023] According to one embodiment of the present invention, the regression equation established is as follows:

[0024] δ hm =b0+b1X1+b2X2+b3X3+b4X4+b5X5+b6X1X2+b7X1X3+b8X1X5 Formula (1)

[0025] Among them, δhm is the maximum wall deflection, b0~b8 are the coefficients of the wall deflection equation;

[0026] X1=t(H e ); X2=t(ln(EI / γ w hav g 4 )); X3=t(B / 2); X4=t(Su / σ′ v ); X5 = t(E i / σ′ v ).

[0027] According to one embodiment of the present invention, the regression equation is determined under the premise of a given excavation depth and data is provided by multiple regression analysis.

[0028] According to one embodiment of the present invention, before performing regression analysis, each of the five input variables is transformed, and the transformation uses the following function:

[0029] X=t(x)=a1x 2 +a2x+a3 Formula (2)

[0030] Where x represents the input variable, selected from H e 、ln(EI / γ w h avg 4 ), B / 2, S u / σ v and E i / σ′ v , X represents the transformed variable.

[0031] According to one embodiment of the present invention, the following three input variables are used in step S3 to establish an empirical equation for predicting the deformation ratio N: clay / H wall 、S u / σ v 、E i / 1000σ′ v ,

[0032] Among them, H clay Indicates the thickness of the clay layer within the steel sheet pile range, H wall Indicates the depth of the steel sheet pile, S u represents the shear strength of soil, σ v ' represents the vertical effective stress, E i represents the initial Young's modulus.

[0033] According to one embodiment of the present invention, the established empirical equation is as follows:

[0034]

[0035] Among them, c0~c8 are coefficients of the deformation ratio equation;

[0036] Y1=∑H clay / H wall ; Y2=S u / σ′ v ; Y3=E i / 1000σ′ v .

[0037] According to one embodiment of the present invention, the machine learning algorithm in step S4 adopts a generalized regression neural network algorithm.

[0038] According to one embodiment of the present invention, the following formula is used in step S5 to determine the maximum surface settlement δ vm :

[0039] δ vm =N·δ hm Formula (4)

[0040] Where N is the deformation ratio, δ hm is the maximum wall deflection.

[0041] According to one embodiment of the present invention, in step S6, the area affected by ground subsidence around the foundation pit is divided into a major impact area, a minor impact area, and a slight impact area.

[0042] According to one embodiment of the present invention, wherein:

[0043] The main impact area meets the following conditions:

[0044]

[0045] The secondary impact area meets the following conditions:

[0046] δ v / δ vm =-0.14×x / h e +0.71 1.5≤x / h e ≤4 The slightly affected area described in formula (6) meets the following conditions:

[0047] δ v / δ vm =-0.0375×x / h e +0.3 4≤x / h e ≤8 Formula (7)

[0048] Among them, x represents the distance between the predicted point and the foundation pit retaining structure, δ v represents the surface settlement of the point to be predicted, δ vm Indicates the maximum surface settlement, he Indicates the excavation depth.

[0049] Another aspect of the present invention provides a ground settlement prediction device for a deep foundation pit in a soft soil area, comprising:

[0050] The surface settlement initial model establishment module is configured to simulate the actual construction conditions of the foundation pit and establish a two-dimensional finite element model of the soil, support, and steel sheet piles for surface settlement analysis under the coupling effect of seepage field and stress field;

[0051] The surface settlement influencing factor determination module is configured to use the control variable method to adjust the working conditions of the established two-dimensional finite element model. Based on the model calculation results, the influencing factors of the ground settlement of the deep foundation pit in the soft soil area are analyzed to determine the main factors affecting the surface settlement;

[0052] a maximum wall deflection determination module configured to establish a regression equation for predicting a maximum wall deflection caused by support excavation based on major factors affecting ground settlement;

[0053] a deformation ratio prediction module configured to establish an empirical equation for predicting a deformation ratio caused by support excavation based on main factors affecting surface settlement;

[0054] A parameter training module is configured to use a machine learning algorithm to train and predict parameters in the regression equation and the empirical equation based on the accumulated surface settlement data obtained through numerical analysis of a typical shaft construction process, and to determine the deflection equation and the deformation ratio equation based on the parameters;

[0055] a maximum surface settlement determination module configured to determine the maximum surface settlement based on a deflection equation and a deformation ratio equation;

[0056] The ground settlement prediction module is configured to divide the ground settlement affected area around the foundation pit into zones based on the maximum surface settlement, the settlement influence range outside the pit, and the shape of the settlement curve, and calculate the surface settlement profile to obtain the surface settlement of any point outside the foundation pit.

[0057] Due to the adoption of the above technical solutions, the method and device for predicting ground settlement of deep shaft foundation pits in soft soil areas provided by the present invention have the following beneficial effects compared with the prior art: the method and device of the present invention simplify the calculation process of ground settlement of shaft foundation pits, overcome the problems of large discreteness of calculation results and strong dependence on personnel in numerical calculation methods, the method comprehensively considers factors such as shaft foundation pit size, enclosure structure, and stratum characteristics, adopts machine learning algorithms to train and predict parameters in empirical formulas, and can quickly calculate ground settlement induced by shaft foundation pit construction, the method can predict the maximum surface settlement and settlement curve caused by the excavation of deep shaft foundation pits in soft soil areas, and then effectively grasp the changes and regularity of soil subsidence during construction, quickly discover problems and take corresponding engineering measures, thereby avoiding possible engineering quality risks and safety hazards. The formula for predicting maximum surface settlement value can provide a certain reference value for future foundation pit projects under soft soil geological conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 is a flow chart of a method for predicting ground settlement of a deep foundation pit in a soft soil area according to one embodiment of the present invention;

[0059] Figure 2 is a mesh division diagram of a foundation pit finite element model according to one embodiment of the present invention;

[0060] Figure 3 is a cloud diagram of vertical displacement of a foundation pit simulated according to an embodiment of the present invention;

[0061] Figure 4 is a diagram showing the effect of excavation depth on surface settlement outside a pit under fluid-solid coupling conditions simulated according to one embodiment of the present invention;

[0062] Figure 5 is a diagram showing the effect of soft soil thickness on surface settlement outside a pit under fluid-solid coupling conditions simulated according to one embodiment of the present invention;

[0063] Figure 6 is a framework diagram of a generalized regression neural network algorithm adopted according to an embodiment of the present invention;

[0064] Figure 7 It is a diagram for dividing the impact of surface settlement outside a foundation pit simulated according to an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0066] One aspect of the present invention provides a method for predicting ground settlement of a deep foundation pit in a soft soil area. Figure 1 As shown, the method generally includes the following steps:

[0067] Step S1: Based on the actual construction simulation conditions of the foundation pit, a two-dimensional finite element model of soil, support, and steel sheet piles for surface settlement analysis is established considering the coupling effect of seepage field and stress field;

[0068] Step S2: Using the control variable method to adjust the working conditions of the two-dimensional finite element model established in step S1, based on the model calculation results, analyze the factors affecting the ground settlement of the deep foundation pit in the soft soil area to determine the main factors affecting the ground settlement;

[0069] Step S3: establishing a regression equation for predicting the maximum wall deflection caused by support excavation and an empirical equation for predicting the deformation ratio based on the main factors affecting surface settlement;

[0070] Step S4: Based on the accumulated surface settlement data obtained through numerical analysis of the typical shaft construction process, a machine learning algorithm is used to train and predict the parameters in the regression equation and the empirical equation, and the deflection equation and the deformation ratio equation are determined based on the parameters;

[0071] Step S5: determining the maximum surface settlement based on the deflection equation and the deformation ratio equation;

[0072] Step S6: Based on the maximum surface settlement, the settlement influence range outside the pit and the shape of the settlement curve, the ground settlement influence area around the foundation pit is divided into zones and the surface settlement profile is calculated to obtain the surface settlement of any point outside the foundation pit.

[0073] This method simplifies the calculation process of ground settlement of vertical shaft foundation pits and overcomes the problems of large discreteness of calculation results and strong dependence on personnel in numerical calculation methods. This method comprehensively considers factors such as the vertical shaft foundation pit size, retaining structure, and stratum characteristics, and uses machine learning algorithms to train and predict parameters in empirical formulas. It can quickly calculate ground settlement induced by vertical shaft foundation pit construction.

[0074] When establishing the two-dimensional finite element model in step S1, the coupling effect of the seepage field and the stress field was taken into account. During the foundation pit dewatering process, the groundwater level drops due to dewatering, and the pore water pressure decreases accordingly. According to the effective stress principle, the decreased pore water pressure value is converted into an effective stress increment. The increase in effective stress causes the soil layer to compact. After the soil layer is compacted, the porosity of the soil changes, causing the permeability coefficient of various parts of the soil to change, affecting its seepage state. Therefore, the ground settlement caused by foundation pit dewatering is a fluid-solid coupling problem. Only by analyzing the coupling effect of the stress field and the seepage field can we better reflect the interaction mechanism between water and soil during the dewatering process and better predict the ground settlement deformation caused by dewatering.

[0075] In step S1, set up a two-dimensional finite element model, which uses the concept of discretization, and divides a continuous medium or engineering structure into a collection of many subdomains of finite size, each small area is called a unit or element, and the collection of units or elements is called a grid. First, study these units or elements, figure out their single mechanical properties, and then assemble all these clearly analyzed units or elements by a certain method, thereby obtaining the understanding of the entire continuous medium or complex engineering structure. In the application of foundation pit engineering, the stress and displacement properties of the supporting structure and the surrounding soil can be analyzed as a whole, and the assembly and disassembly of construction excavation and support can be dynamically simulated, and the influence of seepage, soil consolidation and spatial effect caused by groundwater can be considered simultaneously. In the present invention, specifically finite element analysis software such as MIDAS / GTS, ANSYS, ABAQUS, ADINA can be adopted.

[0076] Figure 2 The mesh division diagram of the finite element model of the foundation pit according to one embodiment of the present invention is shown. The model is established using ABAQUS commercial software, which is used to simulate the ground settlement around a steel sheet pile foundation pit. The steel sheet pile foundation pit is 8m long and 6m wide, with an excavation depth of approximately 8.51m and a retaining structure depth of 15m. The strata at the foundation pit are mainly composed of silty clay and clay. The soil unit in the model adopts the pore pressure unit, that is, the CPE4P plane four-node seepage stress coupling unit, the retaining structure adopts the CPE4 four-node bilinear plane strain quadrilateral unit, and the concrete support and steel support are both simulated using the beam unit B21. The concrete support adopts C35 concrete, and its elastic modulus is E c =31.5GPa, Poisson's ratio is 0.2. The steel support is made of Q235, and its elastic modulus is E t =200GPa, Poisson's ratio is 0.2.

[0077] In some embodiments, the constructed two-dimensional finite element model accounts for ground settlement caused by dewatering and staged excavation. The selected software is used to simulate the entire excavation and dewatering process. The software's seepage calculation function is applied to consider the effects of the coupled seepage and stress fields on the deformation of the retaining structure and the surrounding surface during the excavation and dewatering process. When constructing the two-dimensional finite element model, the initial water level is set at the side boundary of the foundation pit to simulate hydraulic recharge at the boundary. The dewatering process is simulated by setting the water level boundary after the foundation pit is dewatered.

[0078] In some embodiments, when establishing a two-dimensional finite element model, the Geostatic module is used to apply the initial geostress field of the model. The horizontal displacement of the left and right boundaries of the model is limited to 0, the horizontal and vertical displacements of the bottom boundary are limited to 0, and the initial pore pressure of the upper boundary is set to 0. The modified Cambridge constitutive model is used for the upper soil of the model, and the Mohr-Coulomb constitutive model is used for the underlying rock layer. The soil parameters are comprehensively determined based on the measured values ​​of indoor and outdoor tests in the survey report, the derived values ​​of the empirical formula, and the numerical inversion values. The soil unit adopts the pore pressure unit, the enclosure structure adopts the CPE4 four-node bilinear plane strain quadrilateral unit, and the concrete support and steel support both adopt the beam unit B21.

[0079] Figure 3 This is a cloud diagram of the vertical displacement of the foundation pit simulated by using the established two-dimensional finite element model according to an embodiment of the present invention. The figure shows that the surface settlement behind the wall is in a groove shape, with the maximum settlement value of 42 mm about 10 m behind the wall. As the distance away from this point increases, the surface settlement value decreases with increasing distance and gradually approaches zero.

[0080] In step S2, the operating conditions of the established two-dimensional finite element model are adjusted using the control variable method. Based on the model calculation results, the factors affecting the ground settlement of the deep foundation pit in the soft soil area are analyzed to determine the main factors affecting the surface settlement. Specifically, the adjusted operating conditions include, for example, excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness, and soft soil layer thickness. When adjusting the excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness, and soft soil layer thickness, the adjusted variables can be increased in sequence at fixed intervals. Based on the change in surface settlement calculated by the model, it is determined whether the excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness, and / or soft soil layer thickness are used as the main factors affecting the surface settlement.

[0081] In one specific example, the excavation width was increased sequentially from 6m to 8m, then to 10m, and finally to 12m. Based on the model's calculation results, it was found that as the excavation width increased, surface settlement increased significantly, exceeding a set threshold (e.g., 2%). The increase became increasingly pronounced as the excavation width increased. This may be because as the excavation width increased, the steel sheet piles' lateral displacement curves on both sides of the foundation pit caused the width changes to reduce their limiting effect on surface settlement. Based on the surface settlement changes, it can be concluded that excavation width is a major factor affecting surface settlement.

[0082] In a specific embodiment, the excavation depth changes from 12m-14m-16m-18m-20m in sequence, and the excavation depth of the model is changed. Based on the calculation results of the model, it can be seen that the greater the excavation depth, the greater the soil unloading, that is, the disturbance to the soil around the excavation area increases, and the soil pressure difference inside and outside the foundation pit will also increase accordingly, resulting in aggravated deformation of the foundation pit retaining structure. The surface settlement curve outside the pit sinks more and more as the excavation depth of the foundation pit deepens, and the groove-shaped settlement area becomes larger and larger. According to the results of the change in surface settlement, it can be concluded that excavation depth is a major factor affecting surface settlement, and as the excavation depth changes from 12m-14m-16m-18m-20m in sequence, the minimum increase in settlement exceeds the set threshold (for example, 16.7%). Figure 4 The figure shows the influence of excavation depth on surface settlement outside the pit under fluid-solid coupling conditions simulated according to an embodiment of the present invention. As the excavation depth changes from 12m-14m-16m-18m-20m, the greater the excavation depth, the greater the soil unloading, that is, the disturbance to the soil around the excavation area increases, and the soil pressure difference inside and outside the foundation pit also increases, resulting in aggravated deformation of the foundation pit retaining structure. The surface settlement curve outside the pit sinks more and more as the excavation depth of the foundation pit deepens, and the groove-shaped settlement area becomes larger and larger.

[0083] In a specific embodiment, the depth of the steel sheet piles varies from 12m-14m-16m-18m, and the stiffness of the steel sheet piles varies from 10GPa-20GPa-30GPa-40GPa. Based on the calculation results of the model, it can be seen that appropriately increasing the stiffness and embedment depth of the underground steel sheet piles can effectively suppress the surface settlement caused by deep foundation pit excavation. This is because the increase in the stiffness and embedment depth of the steel sheet piles enhances the ability of the retaining structure to resist deformation, thereby limiting the settlement of the surface soil. However, when it increases to a certain extent, the effect of suppressing foundation pit deformation will become very limited. According to the results of the changes in surface settlement, it can be concluded that the depth and stiffness of the steel sheet piles are also the main factors affecting surface settlement. The calculation results of the model show that when the depth of the steel sheet pile changes from 12m-14m-16m in sequence, the minimum change in settlement exceeds the set threshold (for example, 6.7%), but when the depth of the steel sheet pile increases to 18m, the change in settlement does not exceed 1%; when the stiffness of the steel sheet pile changes from 10GPa-20GPa-30GPa in sequence, the minimum change in settlement exceeds the set threshold (for example, 5%), but when the depth of the steel sheet pile increases to 40GPa, the change in settlement does not exceed 1.3%.

[0084] In a specific embodiment, the soft soil thickness increases from 5m-10m-15m-20m-25m-30m. Based on the calculation results of the model, it can be seen that as the thickness of the model soft soil increases from 5m-10m-15m-20m-25m-30m, the concave portion of the curve becomes more obvious, sinks, and the range gradually expands. The reason for this is that soft soil has poor engineering properties such as high compressibility, large thixotropy, low bearing capacity, high sensitivity, and large disturbance. Once dewatering and excavation are carried out, the soft soil is affected, and the soil body undergoes large deformation. Therefore, the higher the proportion of soft soil, the greater the deformation and the range of influence. The calculation results of the model also show that as the thickness of the model soft soil changes from 5m-10m-15m-20m-25m-30m, the minimum change in settlement exceeds the set threshold (for example, 9%). Therefore, the thickness of soft soil is also a major factor affecting surface settlement. Figure 5 A diagram shows the effect of soft soil thickness on surface settlement outside a pit under fluid-solid coupling conditions simulated according to one embodiment of the present invention. As the model soft soil thickness increases from 5m to 10m, 15m to 20m, 25m to 30m, the concave portion of the curve becomes increasingly pronounced, sinking, and its range gradually expanding. This is due to the poor engineering properties of soft soil, such as high compressibility, strong thixotropy, low bearing capacity, high sensitivity, and high perturbability. Once dewatering and excavation occur, the soft soil is affected, resulting in significant deformation.

[0085] In step S3, based on the main factors affecting ground settlement determined in the above steps, a regression equation for predicting the maximum wall deflection caused by support excavation and an empirical equation for predicting deformation ratio are established. First, a regression equation for predicting the maximum wall deflection caused by support excavation is established using five input variables H e , EI / γ w h avg 4 , B, S u / σ′ v and E i / σ v To establish the regression equation for predicting the maximum wall deflection, where H e represents the excavation depth, EI represents the stiffness of the steel sheet pile, γ w Indicates the weight of water, h avg represents the average vertical spacing of horizontal supports, B represents the excavation width, S u represents the shear strength of soil, σ v ' represents the vertical effective stress, E i represents the initial Young's modulus. The values ​​of these variables can be obtained based on the actual project's shaft excavation dimensions, retaining structure design data, and soil shear strength. The influence of hard strata needs to be excluded. A total of 106 representative models were used to establish the predicted maximum wall deflection δ hmThe regression equation is:

[0086]

[0087] Among them, δ hm is the maximum wall deflection, b0~b8 are the coefficients of the wall deflection equation.

[0088] In formula (1),

[0089] X1=t(H e ); X2=t(ln(EI / γ w h avg 4 )); X3 = t(B / 2); X4 = t(S u / σ′ v ); X5 = t(E i / σ′ v ).

[0090] The following assumptions are made in the above steps: (1) the relationship between each of the five variables and the maximum wall deflection is studied under the premise of a given excavation depth; (2) the data is provided by multiple regression analysis. Before the regression analysis, each of the five input variables is transformed to achieve an approximately linear trend between the maximum wall deflection and each of these variables. The form of the transformation function is established through multiple trial and error analysis, with the goal of maintaining a uniform and simple functional form across all input variables. The transformation uses the following function:

[0091] X=t(x)=a1x 2 +a2x+a3 Formula (2)

[0092] Where x represents the input variable, selected from H e 、ln(EI / γ w h avg 4 ), B / 2, S u / σ′ v and E i / σ′ v , X represents the transformed variable.

[0093] Then, an empirical equation for predicting the deformation ratio N is established. The following three input variables are used to establish the empirical equation for predicting the deformation ratio N: clay / H wall 、S u / σ′ v 、E i / 1000σ′ v , where H clay Indicates the thickness of the clay layer within the steel sheet pile range, H wallIndicates the depth of the steel sheet pile, S u represents the shear strength of soil, σ v ' represents the vertical effective stress, E i represents the initial Young's modulus. The values ​​of these variables can be obtained based on the actual project's vertical shaft excavation size, retaining structure design data, soil shear strength and other parameters. The established empirical equation is as follows:

[0094]

[0095] Among them, c0~c8 are coefficients of the deformation ratio equation;

[0096] Y1=∑H clav / H wall ; Y2=S u / σ v ; Y3=E i / 1000σ v .

[0097] Next, in step S4, based on the accumulated surface settlement data obtained from numerical analysis of the typical vertical shaft construction process, a machine learning algorithm is used to train and predict the parameters in the regression equation and the empirical equation. The deflection equation and the deformation ratio equation are then determined based on these parameters. The vertical shaft and the surrounding surface are considered as a whole for this study. Based on an investigation of the factors influencing surface settlement during the typical vertical shaft construction process, a machine learning algorithm is combined with an empirical formula for predicting surface settlement during the typical vertical shaft construction process. The machine learning algorithm is then used to train and predict the parameters in the empirical formula. Figure 6 This is a diagram of the framework of a generalized regression neural network algorithm used in one embodiment of the present invention. The generalized regression neural network uses sample data as a posteriori conditions and calculates the network output according to the maximum probability principle. The generalized regression neural network algorithm is used to train and predict parameters in regression equations and empirical equations. Table 1 below lists the coefficients of the wall deflection equation obtained through training according to one embodiment of the present invention, and Table 2 lists the coefficients of the deformation ratio equation obtained through training according to one embodiment of the present invention.

[0098] Table 1 Wall deflection equation coefficients

[0099]

[0100] Table 2 Deformation ratio equation coefficients

[0101]

[0102] Substitute each coefficient value into the corresponding wall deflection equation and deformation ratio equation. Next, in step S5, the maximum surface settlement is determined based on the determined deflection equation and deformation ratio equation. The calculation formula for the maximum surface settlement δvm is as follows:

[0103] δvm =N·δ hm Formula (4)

[0104] Where N is the deformation ratio, δ hm is the maximum wall deflection.

[0105] After determining the maximum surface settlement δvm, in step S6, based on the maximum surface settlement and the influence range of the settlement outside the pit and the shape of the settlement curve, the ground settlement influence area around the foundation pit can be divided and the surface settlement profile can be calculated to obtain the surface settlement of any point outside the foundation pit. According to the existing paper results and actual working conditions, the influence range of the settlement outside the pit can be determined. Combined with the simulation results of the previous model and the shape of the settlement curve, the influence range can be divided into the main influence area, the secondary influence area and the slight influence area. The surface settlement in the main influence area is mainly caused by excavation, and the surface settlement in the secondary influence area is mainly caused by precipitation. Combined with some measured surface settlement data (mainly finite element model simulation data), the surface settlement calculation formula of each partition is determined by fitting. Each partition can be calculated by the following formula:

[0106] The main impact area meets the following conditions:

[0107]

[0108] The secondary impact area meets the following conditions:

[0109] δ v / δ vm =-0.14×x / h e +0.71 1.5≤x / h e ≤4 Formula (6)

[0110] The slightly affected area meets the following conditions:

[0111] δ v / δ vm =-0.0375×x / h e +0.3 4≤x / he≤8 Formula (7)

[0112] Among them, x represents the distance between the predicted point and the foundation pit retaining structure, δ v represents the surface settlement of the point to be predicted, δ vm Indicates the maximum surface settlement, h e Indicates the excavation depth. Based on x and h e Select the appropriate formula for the ratio, and then according to the formula and the corresponding δ vm The surface settlement at any point outside the foundation pit can be calculated. Figure 7The figure shows the division of the impact of surface settlement outside the foundation pit simulated according to an embodiment of the present invention. The impact range of the surrounding surface settlement caused by dewatering of the deep foundation pit in the soft soil area is divided into three impact areas: major, minor and slight.

[0113] This method can predict the maximum surface settlement and settlement curve caused by deep vertical foundation pit excavation in soft soil areas. This method effectively captures the dynamics and regularity of soil subsidence during construction, allowing for rapid identification of problems and implementation of appropriate engineering measures to mitigate potential project quality risks and safety hazards. The formula for predicting maximum surface settlement accurately predicts maximum surface settlement and provides valuable insights for future foundation pit projects in soft soil conditions.

[0114] Another aspect of the present invention provides a device for predicting ground settlement of a deep foundation pit in a soft soil area, the device comprising:

[0115] The surface settlement initial model establishment module is configured to simulate the actual construction conditions of the foundation pit and establish a two-dimensional finite element model of the soil, support, and steel sheet piles for surface settlement analysis under the coupling effect of seepage field and stress field;

[0116] The surface settlement influencing factor determination module is configured to use the control variable method to adjust the working conditions of the established two-dimensional finite element model. Based on the model calculation results, the influencing factors of the ground settlement of the deep foundation pit in the soft soil area are analyzed to determine the main factors affecting the surface settlement;

[0117] a maximum wall deflection determination module configured to establish a regression equation for predicting a maximum wall deflection caused by support excavation based on major factors affecting ground settlement;

[0118] a deformation ratio prediction module configured to establish an empirical equation for predicting a deformation ratio caused by support excavation based on main factors affecting surface settlement;

[0119] A parameter training module is configured to use a machine learning algorithm to train and predict parameters in the regression equation and the empirical equation based on the accumulated surface settlement data obtained through numerical analysis of a typical shaft construction process, and to determine the deflection equation and the deformation ratio equation based on the parameters;

[0120] a maximum surface settlement determination module configured to determine the maximum surface settlement based on a deflection equation and a deformation ratio equation;

[0121] The ground settlement prediction module is configured to divide the ground settlement affected area around the foundation pit into zones based on the maximum surface settlement, the settlement influence range outside the pit, and the shape of the settlement curve, and calculate the surface settlement profile to obtain the surface settlement of any point outside the foundation pit.

[0122] The above-described embodiments merely illustrate the implementation methods of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make various modifications and improvements without departing from the scope of the present invention, and all such modifications and improvements fall within the scope of protection of the present invention.

Claims

1. A method for predicting ground settlement of a deep foundation pit in a soft soil area, characterized in that: The following steps are involved: Step S1: Based on the actual construction simulation conditions of the foundation pit, a two-dimensional finite element model of soil, support, and steel sheet piles for surface settlement analysis is established considering the coupling effect of seepage field and stress field; Step S2: Using the control variable method to adjust the working conditions of the two-dimensional finite element model established in step S1, based on the model calculation results, analyze the factors affecting the ground settlement of the deep foundation pit in the soft soil area to determine the main factors affecting the ground settlement; Step S3: establishing a regression equation for predicting the maximum wall deflection caused by support excavation and an empirical equation for predicting the deformation ratio based on the main factors affecting surface settlement; Step S4: Based on the accumulated surface settlement data obtained through numerical analysis of the typical shaft construction process, a machine learning algorithm is used to train and predict the parameters in the regression equation and the empirical equation, and the deflection equation and the deformation ratio equation are determined based on the parameters; Step S5: determining the maximum surface settlement based on the deflection equation and the deformation ratio equation; Step S6: Based on the maximum surface settlement, the settlement influence range outside the pit and the settlement curve shape, the ground settlement influence area around the foundation pit is divided into zones and the surface settlement profile is calculated to obtain the surface settlement of any point outside the foundation pit.

2. The method according to claim 1, characterized in that When establishing the two-dimensional finite element model in step S1, the ground settlement caused by precipitation and step-by-step excavation is taken into account.

3. The method according to claim 2, characterized in that When establishing the two-dimensional finite element model in step S1, the initial water level is set at the side boundary of the foundation pit to simulate the hydraulic recharge of the boundary, and the precipitation process is simulated by setting the water level boundary after the foundation pit is dewatered.

4. The method according to claim 1, wherein When establishing the two-dimensional finite element model in step S1, the Geostatic module is used to apply the initial geostress field of the model.

5. The method according to claim 1, wherein In step S1, the horizontal displacement of the left and right boundaries of the model is restricted to 0, the horizontal and vertical displacements of the bottom boundary are restricted to 0, and the initial pore pressure of the upper boundary is set to 0.

6. The method according to claim 1, characterized in that In step S1, the modified Cambridge constitutive model is used for the upper soil of the model, and the Mohr-Coulomb constitutive model is used for the underlying rock layer. The soil parameters are comprehensively determined based on the measured values ​​of indoor and outdoor tests in the survey report, the derived values ​​of empirical formulas, and the numerical inversion values. The soil element uses the pore pressure element, the enclosure structure uses the CPE4 four-node bilinear plane strain quadrilateral element, and the concrete support and steel support use the beam element B21.

7. The method according to claim 1, characterized in that The working conditions adjusted in step S2 include excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness and soft soil layer thickness.

8. The method according to claim 7, characterized in that In step S2, when adjusting the excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness and soft soil layer thickness, the adjusted variables are sequentially increased at a fixed interval, and the change in surface settlement calculated by the model is used to determine whether the excavation width, excavation depth, steel sheet pile depth, steel sheet pile stiffness and / or soft soil layer thickness are used as the main factors affecting the surface settlement.

9. The method according to claim 1, characterized in that In step S3, the following five input variables are used to establish a regression equation for predicting the maximum wall deflection: H e , EI / γ w h avg 4 , B, S u / σ′ v and E i / σ′ v , where H e represents the excavation depth, EI represents the stiffness of the steel sheet pile, γ w Indicates the weight of water, h avg represents the average vertical spacing of horizontal supports, B represents the excavation width, S u represents the shear strength of soil, σ v ' represents the vertical effective stress, E i represents the initial Young's modulus.

10. The method according to claim 9, characterized in that The established regression equation is as follows: δ hm = b0 + b1X1 + b2X2 + b3X3 + b4X4 + b5X5 + b6X1X2 + b7X1X3 + b8X1X5 Equation (1) Among them, δhm is the maximum wall deflection, b0~b8 are the coefficients of the wall deflection equation; X1=t(H e );X2=t(ln(EI / γ w h avg 4 ));X3=t(B / 2);X4=t(S u / σ′ v );X5=t(E i / σ′ v )。 11. The method according to claim 9, characterized in that The regression equation is determined under the premise of a given excavation depth and the data are provided by multiple regression analysis.

12. The method according to claim 9, characterized in that Before performing the regression analysis, each of the five input variables was transformed using the following function: X=t(x)=a1x 2 +a2x+a3 Formula (2) Where x represents the input variable, selected from H e 、ln(EI / γ w h avg 4 ), B / 2, S u / σ′ v and E i / σ′ v , X represents the transformed variable.

13. The method according to claim 1, wherein In step S3, the following three input variables are used to establish an empirical equation for predicting the deformation ratio N: clay / H wall 、S u / σ′ v 、E i / 1000σ′ v , where H clay Indicates the thickness of the clay layer within the steel sheet pile range, H wall Indicates the depth of the steel sheet pile, S u represents the shear strength of soil, σ v ' represents the vertical effective stress, E i represents the initial Young's modulus.

14. The method according to claim 13, characterized in that The established empirical equation is as follows: Among them, c0~c8 are coefficients of the deformation ratio equation; Y1=∑H clay / H wall ;Y2=S u / s′ v ;Y3=E i / 1000σ′ v 。 15. The method according to claim 1, wherein The machine learning algorithm in step S4 adopts a generalized regression neural network algorithm.

16. The method according to claim 1, wherein In step S5, the following formula is used to determine the maximum surface settlement δ vm : d vm =N·δ hm expression(4) Where N is the deformation ratio, δ hm is the maximum wall deflection.

17. The method according to claim 1, wherein In step S6, the land subsidence affected area around the foundation pit is divided into a major impact area, a minor impact area and a slight impact area.

18. The method according to claim 17, characterized in that in: The main impact area meets the following conditions: The secondary impact area meets the following conditions: d v / d vm =-0.14×x / h e +0.71 1.5≤x / h e ≤4 expressions(6) The slightly affected area meets the following conditions: d v / d vm =-0.0375×x / h e +0.3 4≤x / h e ≤8 expressions(7) Among them, x represents the distance between the predicted point and the foundation pit retaining structure, δ v represents the surface settlement of the point to be predicted, δ vm Indicates the maximum surface settlement, h e Indicates the excavation depth.

19. A ground settlement prediction device for a deep foundation pit in a soft soil area, characterized in that: include: The surface settlement initial model establishment module is configured to simulate the actual construction conditions of the foundation pit and establish a two-dimensional finite element model of the soil, support, and steel sheet piles for surface settlement analysis under the coupling effect of seepage field and stress field; The surface settlement influencing factor determination module is configured to use the control variable method to adjust the working conditions of the established two-dimensional finite element model. Based on the model calculation results, the influencing factors of the ground settlement of the deep foundation pit in the soft soil area are analyzed to determine the main factors affecting the surface settlement; a maximum wall deflection determination module configured to establish a regression equation for predicting a maximum wall deflection caused by support excavation based on major factors affecting ground settlement; a deformation ratio prediction module configured to establish an empirical equation for predicting a deformation ratio caused by support excavation based on main factors affecting surface settlement; A parameter training module is configured to use a machine learning algorithm to train and predict parameters in the regression equation and the empirical equation based on the accumulated surface settlement data obtained through numerical analysis of a typical shaft construction process, and to determine the deflection equation and the deformation ratio equation based on the parameters; a maximum surface settlement determination module configured to determine the maximum surface settlement based on a deflection equation and a deformation ratio equation; The ground settlement prediction module is configured to divide the ground settlement affected area around the foundation pit into zones based on the maximum surface settlement, the settlement influence range outside the pit, and the shape of the settlement curve, and calculate the surface settlement profile to obtain the surface settlement of any point outside the foundation pit.

Images