Method for evaluating safety state of tunnel lining with cracks

Abstract

The invention discloses a crack-containing tunnel lining safety state evaluation method. The method comprises the following steps: S1, calculating a whole-ring safety coefficient of a lining structure when no crack exists; s2, calculating the influence coefficient of the single crack on the safety state of the structure under different crack characteristics; s3, performing partition definition on the dendritic crack by using the branch part and the branch part, and calculating the influence coefficients of the branch part and the branch part on the structure safety state under different crack characteristics; s4, constructing and training a neural network model, and training the neural network model by using the data obtained through simulation calculation in S2 and S3; and S5, segmenting the crack along the crack trend, calculating an influence coefficient of each segment according to the neural network model trained in the step S4, and calculating a residual safety coefficient of each segment to obtain residual safety coefficient distribution. According to the method, the safety influence of the cracks in different positions of the tunnel lining structure on the tunnel lining structure can be efficiently and accurately judged.

Description

Technical Field

[0001] This invention relates to the field of tunnel condition assessment technology, and in particular to a method for assessing the safety condition of tunnel lining with cracks. Background Technology

[0002] Due to the complex and variable service environment of railway tunnels, an increasing number of tunnels are experiencing lining defects. Lining cracks, as one of the common defects in tunnel linings, can seriously affect the overall structural stability of the tunnel lining. Therefore, establishing an objective and scientific evaluation system for cracked tunnel lining structures has become a key issue in the field of tunnel condition assessment.

[0003] Traditional analysis of the impact of apparent cracks mainly uses numerical simulation methods. However, existing methods for simulating cracks in tunnel lining structures have several shortcomings: First, given that two-dimensional simulations are used to analyze the safety impact of cracks on the lining, it is difficult to reflect the three-dimensional stress state at the crack location, leading to inaccurate analysis conclusions. Second, conventional three-dimensional analysis often uses the entire ring or section as the analysis object, but the modeling or calculation process is still relatively complex and time-consuming, which directly makes it difficult to conduct a scientific and efficient analysis of the hazards of cracks.

[0004] In view of this, this invention is hereby proposed. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method for assessing the safety status of tunnel linings with cracks. Compared to two-dimensional analysis, this method can reflect the three-dimensional stress state and achieve more accurate analysis of the impact of crack damage. Compared to conventional three-dimensional analysis, by establishing the locally equivalent simplified model proposed in this invention, it can efficiently and accurately simulate and analyze the safety impact of cracks located at different positions in the full-ring lining structure of railway tunnels on the tunnel lining structure.

[0006] To achieve the above objectives, the present invention also employs the following technical solution: A method for assessing the safety status of tunnel lining with cracks includes the following steps: S1: Using the load-structure method and combining the actual design parameters of the tunnel to be evaluated, calculate the full-ring safety factor of the lining structure when there are no cracks. S2: For single-crack defects, a local equivalent simplified mathematical model and corresponding numerical simulation model of tunnel lining structure containing single cracks are established, and the influence coefficient of single crack on structural safety state under different crack characteristics is calculated. S3: For dendritic crack defects, the dendritic cracks are defined by branching parts and forked parts. For branching parts, a local equivalent simplified model of the dendritic crack is established, and the ultimate bearing capacity curve is plotted by applying different combinations of bending moment and axial force to the model. Then, the influence coefficient of the branching part of the dendritic crack is obtained. The calculation of the forked part uses the calculation method of single crack defects in step S2. A local equivalent simplified mathematical model and a corresponding numerical simulation model of the tunnel lining structure are established respectively. Then, the influence coefficient of the forked part of the dendritic crack is obtained. The influence coefficient of the forked part and the branching part on the structural safety state under different crack characteristics is calculated. S4: Construct and train a neural network model with crack geometry as input and influence coefficient as output, and use the data obtained from the simulation calculations in S2 and S3 to train the neural network model; S5: Divide the actual cracks into segments along the crack direction, extract the location of the cracks in different segments in the entire tunnel ring, calculate the influence coefficient of each segment based on the neural network model trained in S4, and calculate the remaining safety factor of each segment to obtain the distribution of the remaining safety factor.

[0007] Furthermore, step S1 includes the following steps: S11: Establish a tunnel lining simulation model. Based on the characteristics of the tunnel lining and the actual situation, establish an ABAQUS finite element simulation model of the full-ring tunnel lining. S12: Set the surrounding rock pressure. Based on the calculation method of surrounding rock pressure in the railway tunnel design code, calculate the surrounding rock pressure for a general deep-buried tunnel to obtain design parameters such as vertical surrounding rock pressure and horizontal surrounding rock pressure. S13: Determine the minimum compressive stress value of the entire ring, extract the stress distribution through numerical simulation, and compare it with the railway tunnel design specifications to obtain the minimum stress value of the entire ring of the crack-free tunnel lining; S14: Calculate the safety factor without cracks. Based on the calculation method of structural safety factor in the railway tunnel design code, the safety factor of the above tunnel structure is calculated using the damage stage method.

[0008] Furthermore, step S2 includes the following steps: S21: The actual complete tunnel lining crack is equivalent to a local equivalent model. A three-dimensional reinforced concrete local equivalent simplified model of the single-crack tunnel lining structure is established. The reinforced concrete local equivalent simplified model includes a concrete model and a steel reinforcement model. A Cartesian coordinate system is defined. The model shape is: a concrete slab at the center, two concrete fixing members in the x and y directions of the concrete slab, and a total of four concrete fixing members. Each concrete fixing member is connected to the middle concrete slab by a spring. S22: Set the model material parameters. The material parameters of the concrete model and the steel reinforcement model should be consistent with the tunnel lining material parameters. S23: Set the model load and set uniformly distributed compressive stress on the two sides of the concrete slab along the y direction. The direction of the uniformly distributed pressure is the center of the concrete slab, and the magnitude is the minimum compressive stress value of the whole ring determined in step S13. S24: Set the model boundary. Set the completely fixed boundary for the face of the four fixed concrete components that is in the opposite direction to the face that connects to the middle concrete slab. S25: Set a single crack in the model, establish a precast crack at the center of the lower concrete slab surface based on the crack characteristics, and use the extended finite element module to set the precast crack as a crack, set the entire concrete slab as the extended finite element crack propagation area, and allow crack growth. S26: Using the static analysis in the ABAQUS analysis step module, apply compressive stress to the center of the surface of the concrete slab in the positive z-axis direction until the concrete fails, simulate the entire process of the propagation of a precast single crack in the concrete, and calculate the ultimate bearing capacity of the tunnel lining. S27: The influence coefficient is calculated using the following formula: ; In the formula, α is the influence coefficient, F is the ultimate bearing capacity of the local equivalent simplified model of tunnel lining under single crack, and F0 is the ultimate bearing capacity of the local equivalent simplified model of tunnel lining without crack.

[0009] Furthermore, in S24, spring units are used to simulate the constraint effect between a local area of ​​the tunnel lining and the surrounding structure; The spring stiffness is determined using the following formula: K = K 0×(0.197+0.803 e -20.426h / H ); In the formula, K 0 represents the original stiffness of the concrete structure, which is determined by the ratio of the concrete strength to the number of springs. h The depth of the crack. H The thickness of the concrete structure; The stiffness of the spring element is taken as the stiffness of the concrete or reinforced concrete when the crack penetrates; the spring elements are arranged on the four sides of the concrete slab opposite to the four concrete fixing members, one end of the spring element is connected to the concrete slab and the other end is connected to the concrete fixing members; the spring elements are used to constrain the concrete slab in three orthogonal directions.

[0010] Furthermore, in S25, the crack angle is defined by the angle between the crack length direction and the tunnel longitudinal direction; the crack depth is defined by the maximum distance the crack extends from the lower surface of the concrete slab along the z-axis in a direction perpendicular to the surface into the interior of the concrete slab; and the crack direction is defined by the extension direction of the crack along the crack surface in terms of both angle and depth. The crack angles are set to 0°, 10°, 20°, 30°, 40°, 50°, 60°, 70°, 80°, and 90°, respectively; and the crack depths are set to 0.05d, 0.1d, 0.2d, 0.3d, 0.5d, 0.7d, 0.8d, and 0.9d, respectively, where d is the original tunnel lining thickness. These three parameters are combined to form the input parameters for the simulation. Simulation calculations are performed in S26, and influence coefficients are calculated in S27.

[0011] Furthermore, S3 performs simulation calculations on the branched portion through the following steps: S31: Establish a locally equivalent simplified mathematical model of tunnel lining based on crack bifurcation morphology and crack depth. The dendritic crack morphology is represented by the combination of the angle between each crack in the branch and the tunnel circumferential direction. The locally equivalent simplified mathematical model is divided into a steel reinforcement model and a concrete model containing dendritic cracks. Define a Cartesian coordinate system and the model shape is a cuboid concrete block. S32: Set the model material parameters. The material parameters of the concrete model and the steel reinforcement model should be consistent with the tunnel lining material parameters. S33: Set the model boundary, setting a completely fixed boundary for all nodes on the lower surface of the model in the z-direction; S34: Set the dendritic cracks in the model; S35: Using the explicit dynamic analysis in the ABAQUS analysis step module, a combination of bending moment and axial force is applied to the center of the upper surface of the model along the positive z-axis until the model fails, simulating the entire process of the precast branch section's concrete expansion. First, a compressive load is applied to the upper surface of the locally equivalent simplified model of the branch section along the positive z-axis, resulting in point (0, N). Then, the ordinate of point (0, N) is divided into equal parts, and a bending moment is applied again with a fixed axial force value for one of the divisions until the locally equivalent simplified model of the branch section fails. The bending moment value at this point is recorded, and the combination of internal forces (M, N) at this point is taken as a point on the ultimate bearing capacity curve. Finally, the axial force is set to a fixed value of zero, and a pure bending moment is applied until the locally equivalent simplified model of the branch section fails. The bending moment value at this point is recorded, and the combination of internal forces (M, 0) at this point is the intersection of the ultimate bearing capacity curve and the horizontal axis, from which the ultimate bearing capacity curve of the substructure can be plotted. The influence coefficient can be determined by calculating the ratio of the area of ​​the ultimate bearing capacity curve under the condition of the locally equivalent simplified model without cracks to the area of ​​the ultimate bearing capacity curve under the condition of the locally equivalent simplified model with branch cracks.

[0012] S36: The influence coefficient is calculated using the following formula: ; In the formula, α is the influence coefficient, and S u S represents the area of ​​the ultimate bearing capacity curve of the locally equivalent simplified model of the tunnel lining with branched cracks, while S represents the area of ​​the ultimate bearing capacity curve of the locally equivalent simplified model of the tunnel lining without cracks.

[0013] Furthermore, in S34, the crack angle is defined by the angle between the crack length direction and the tunnel longitudinal direction; the crack depth is defined by the maximum distance the crack extends from the lower surface of the concrete slab along the z-axis in a direction perpendicular to the surface into the interior of the concrete slab; and the crack direction is defined by the extension direction of the crack along the crack surface in terms of angle and depth. The crack angles are set to 10°, 20°, 30°, 40°, 60°, 70°, 80°, and 90°, respectively; and the crack depths are set to 0.05d, 0.1d, 0.3d, 0.5d, 0.7d, and 0.9d, respectively, where d is the original tunnel lining thickness. These three parameters are combined to form the input parameters for the simulation. Simulation calculations are performed in S35, and influence coefficients are calculated in S36.

[0014] Furthermore, the neural network model is a BP neural network model with three layers: an input layer, a hidden layer, and an output layer. The inputs are the crack depth and crack angle, and the output is the influence coefficient.

[0015] Furthermore, in step S5, the remaining safety factor for each segment is calculated using the following formula: s i =S i ×α; In the formula, s i S is the remaining safety factor. i α represents the safety factor for the crack-free lining structure across the entire ring, and α is the influence factor.

[0016] Furthermore, in S5, the remaining safety factor is classified based on the calculation results: Level I: 0≤si<2 is very dangerous, the tunnel lining structure is almost ineffective; Level II: 2≤si<6 is dangerous, the tunnel lining structure has a considerable probability of failure; Level III: 6≤si<10 is relatively safe, the tunnel lining structure functions normally and is difficult to fail; Level IV: 10≤si is safe, the tunnel lining structure functions normally.

[0017] Compared with the prior art, the beneficial effects of this invention are as follows: 1. In section S2, for single-crack defects, a locally equivalent simplified model considering crack depth, angle, and structural stress characteristics is proposed. By calculating the change in the ultimate bearing capacity of the locally equivalent simplified model with and without cracks, the influence coefficient of a single crack on the tunnel safety state is determined. In section S3, for dendritic crack defects, the "branch" and "forked" regions of the crack are defined. For the "forked" region, the influence coefficient can be determined using the calculation method for the impact of single-crack defects on the safety state. For the "forked" part, a locally equivalent simplified model of dendritic crack considering crack depth and angle is established. A compressive load is applied to the positive z-axis surface of the locally equivalent simplified model of the branch part, thus obtaining the point (0, N). Then, the ordinate of the point (0, N) is divided into equal parts, and a bending moment is applied with a certain axial force value as a constant until the locally equivalent simplified model of the branch part fails. The result is recorded at this point. The bending moment value is then used as a point on the ultimate bearing capacity curve, where the internal force combination (M, N) is taken as the point. Finally, the axial force is set to a constant value of zero, and then a pure bending moment is applied until the local equivalent simplified model of the branch section fails. The bending moment value is recorded at this point, and the internal force combination (M, 0) is the intersection of the ultimate bearing capacity curve and the horizontal axis. The ultimate bearing capacity curve of the substructure can be plotted. The ratio of the area of ​​the ultimate bearing capacity curve under the condition of the crack-free local equivalent simplified model to the area of ​​the ultimate bearing capacity curve under the condition of the branch crack local equivalent simplified model can be used to determine the influence coefficient. In S4, a neural network model of the tunnel safety state influence coefficient characterized by the crack geometry is established. In S5, for actual cracks, the product of the influence coefficient and the safety factor of the crack-free lining is used as the structural residual safety factor, and the safety state level is evaluated to realize the calculation of the structural residual safety factor within the crack distribution range at different locations.

[0018] 2. Compared with traditional two-dimensional analysis, this invention can reflect the three-dimensional stress state and achieve more accurate analysis of the impact of cracks. Compared with conventional three-dimensional analysis, this invention can efficiently and accurately simulate and analyze the impact of cracks located at different locations in the full ring lining structure of railway tunnels on the safety of the tunnel lining structure. Attached Figure Description

[0019] Figure 1 A flowchart of a method for assessing the safety status of tunnel lining with cracks; Figure 2 Map showing the distribution of local defects in the tunnel Figure 3 A schematic diagram showing the safety factor distribution of a crack-free tunnel lining. Figure 4 This is a schematic diagram showing the markings in the complete model and the local equivalent model of a single-crack tunnel lining structure. Figure 5 A schematic diagram of the boundary and load of a locally equivalent simplified model of a single-crack tunnel lining structure; Figure 6 A schematic diagram of a spring in a locally equivalent simplified model of a single-crack tunnel lining. Figure 7 A schematic diagram of crack propagation in a locally equivalent simplified model of a single-crack tunnel lining. Figure 8 Typical failure diagram of the lower surface of a locally equivalent simplified model of a single-crack tunnel lining; Figure 9 A schematic diagram of a locally equivalent simplified model of the branched portion of the branched crack; Figure 10 A schematic diagram of crack propagation in the branch section of the branch crack, based on numerical simulation.

[0020] Figure 11 Schematic diagram of crack subdivision Figure 12 Schematic diagram of the remaining safety factor distribution within the crack subdivision unit Figure 13 Schematic diagram of the remaining safety factor classification within the crack subdivision unit Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0022] Example 1: A method for assessing the safety status of tunnel lining with cracks, such as... Figure 1 As shown, it includes the following steps: S1: Using the load-structure method and combining the actual design parameters of the tunnel to be evaluated, calculate the full-ring safety factor of the lining structure when there are no cracks. S2: For single-crack defects, a local equivalent simplified mathematical model and corresponding numerical simulation model of tunnel lining structure containing single cracks are established, and the influence coefficient of single crack on structural safety state under different crack characteristics is calculated. S3: For dendritic crack defects, the dendritic cracks are defined by branching parts and forked parts. For branching parts, a local equivalent simplified model of the dendritic crack is established, and the ultimate bearing capacity curve is plotted by applying different combinations of bending moment and axial force to the model. Then, the influence coefficient of the branching part of the dendritic crack is obtained. The calculation of the forked part uses the calculation method of single crack defects in step S2. A local equivalent simplified mathematical model and a corresponding numerical simulation model of the tunnel lining structure are established respectively. Then, the influence coefficient of the forked part of the dendritic crack is obtained. The influence coefficient of the forked part and the branching part on the structural safety state under different crack characteristics is calculated. S4: Construct and train a neural network model with crack geometry as input and influence coefficient as output, and use the data obtained from the simulation calculations in S2 and S3 to train the neural network model; S5: Divide the actual cracks into segments along the crack direction, extract the location of the cracks in different segments in the entire tunnel ring, calculate the influence coefficient of each segment based on the neural network model trained in S4, and calculate the remaining safety factor of each segment to obtain the distribution of the remaining safety factor.

[0023] The safety status assessment method for cracked tunnel lining in this embodiment includes the following steps: In S1, the load-structure method is used to calculate the full-ring safety factor of crack-free lining; In S2, for single-crack defects, a locally equivalent simplified model considering crack depth, angle, and structural stress characteristics is proposed. By calculating the change in the ultimate bearing capacity of the locally equivalent simplified model with and without cracks, the influence coefficient of a single crack on the tunnel safety status is determined; In S3, for branched crack defects, the "branch" and "forked" regions of the crack are defined. For the "forked" region of the crack, the influence coefficient can be determined using the calculation method for the impact of single-crack defects on the safety status. For the "forked" part of the crack, a locally equivalent simplified model of branched crack considering crack depth and crack angle is established. A compressive load is applied to the positive z-axis surface of the locally equivalent simplified model of the branched part to obtain the point (0, N); then the ordinate of the point (0, N) is divided into equal parts, and a bending moment is applied with the axial force value of a certain division as a constant. The process continues until the locally equivalent simplified model of the branch section fails, recording the bending moment value at this point. The internal force combination (M, N) at this point is then taken as a point on the ultimate bearing capacity curve. Finally, the axial force is set to a constant value of zero, and then pure bending moment is applied until the locally equivalent simplified model of the branch section fails. The bending moment value at this point is recorded, and the internal force combination (M, 0) at this point is the intersection of the ultimate bearing capacity curve and the horizontal axis. The ultimate bearing capacity curve of the substructure can be plotted. The ratio of the area of ​​the ultimate bearing capacity curve under the crack-free locally equivalent simplified model condition to the area of ​​the ultimate bearing capacity curve under the branch crack locally equivalent simplified model condition can be used to determine the influence coefficient. In S4, a neural network model of the tunnel safety state influence coefficient characterized by crack geometry is established. In S5, for actual cracks, the product of the influence coefficient and the safety factor of the crack-free lining is used as the structural residual safety factor, and the safety state level is evaluated to realize the calculation of the structural residual safety factor within the crack distribution range at different locations. Compared with existing technologies, the safety status assessment method for tunnel lining with cracks in this embodiment can scientifically analyze the impact of cracks on the tunnel lining structure, improving the efficiency and accuracy of the analysis. By segmenting the cracks and calculating their impact on the tunnel lining structure, the accuracy and efficiency are improved, providing a scientific basis for the quantitative assessment of the impact of cracks on tunnel lining.

[0024] The following describes the specific implementation process of the method in this embodiment, using the actual inspection results of a railway tunnel as background. This line is a passenger and freight line designed for a speed of 200 km / h. Construction of the tunnel began in April 2003, was completed in December 2004, and opened to traffic in November 2005. The tunnel is a single-bore, double-track tunnel. The analyzed section is a Class IV surrounding rock reinforced section, where the secondary lining uses C30 concrete with a thickness of 45 cm. The tunnel is located in a hilly terrain with a relative elevation difference of 150 m. The surface soil is Quaternary colluvial and colluvial silty clay, 2-10 m thick, with a thicker layer of topsoil at the entrance. The bedrock is mainly composed of mudstone interbedded with sandstone and sandstone from the Middle Jurassic Upper Shaximiao Formation. The strata are single-layered, with gently sloping rock layers, well-developed joints, and poor groundwater development. However, the thick sandstone layer has a water-retaining structure. Local cracks and defects are distributed in the tunnel, such as... Figure 2 As shown.

[0025] In an optional embodiment, step S1 includes the following steps: S11: Establish a tunnel lining simulation model. Based on the characteristics of the tunnel lining and the actual situation, establish an ABAQUS finite element simulation model of the full-ring tunnel lining. In one optional embodiment, based on the characteristics of the tunnel lining and the actual situation, an ABAQUS finite element simulation model of the full-ring tunnel lining is established, and the longitudinal length of the model is set to 1m.

[0026] S12: Set the surrounding rock pressure. Based on the calculation method of surrounding rock pressure in the railway tunnel design code, calculate the surrounding rock pressure for a general deep-buried tunnel to obtain design parameters such as vertical surrounding rock pressure and horizontal surrounding rock pressure. In this optional embodiment, based on the rock pressure calculation method in the "Railway Tunnel Design Code" (TB10003-2016), the rock pressure is calculated for a typical deep-buried tunnel, with a vertical pressure of 104.933 kN / m. 2 Horizontal pressure 26.233 kN / m 2 The surrounding rock pressure was applied by surface load, and the surrounding rock resistance was simulated by grounding spring. The elastic resistance coefficient was taken as 200MPa / m.

[0027] S13: Determine the minimum compressive stress value of the entire ring, extract the stress distribution through numerical simulation, and compare it with the railway tunnel design specifications to obtain the minimum stress value of the entire ring of the crack-free tunnel lining; In this optional embodiment, stress distribution is extracted through numerical simulation and compared with the "Railway Tunnel Design Code" to obtain the minimum stress value of the entire ring of the crack-free tunnel lining.

[0028] S14: Calculate the safety factor without cracks. Based on the calculation method of structural safety factor in the railway tunnel design code, the safety factor of the above tunnel structure is calculated using the damage stage method. In this optional embodiment, the safety factor of the tunnel structure is calculated using the failure stage method, based on the calculation method for structural safety factors in the "Railway Tunnel Design Code". Figure 3 As shown, the distribution of the safety factor for crack-free conditions is obtained.

[0029] In an optional embodiment, step S2 includes the following steps: S21: The actual complete tunnel lining crack is equivalent to a local equivalent model. A three-dimensional reinforced concrete local equivalent simplified model of the single-crack tunnel lining structure is established. The reinforced concrete local equivalent simplified model includes a concrete model and a steel reinforcement model. A Cartesian coordinate system is defined. The model shape is: a concrete slab at the center, two concrete fixing members in the x and y directions of the concrete slab, and a total of four concrete fixing members. Each concrete fixing member is connected to the middle concrete slab by a spring. In this optional embodiment, the actual complete single-crack tunnel lining is equivalent to a locally equivalent model, such as... Figure 4 As shown, ignoring the influence of tunnel lining curvature, the local equivalent model is further simplified into a concrete model and a reinforcement model. Using the extended finite element module in ABAQUS software, a three-dimensional locally equivalent simplified concrete model of a single-crack tunnel lining structure is established. Considering the reinforcement arrangement and the thickness of the protective layer, the three-dimensional parameters of the concrete slab are established as follows: length l =2.362m, width w =2.362m, height t =0.45m, the three-dimensional parameters of the concrete fixed member are: length l = 2.362m, width w =0.5m, height t =0.45m.

[0030] S22: Set the model material parameters. The material parameters of the concrete model and the steel reinforcement model should be consistent with the tunnel lining material parameters. In this optional embodiment, the mechanical parameters of the concrete material are set using the properties module in ABAQUS software, and a crack initiation criterion for damaged materials is defined. The concrete material is C30 concrete, and its material properties are set to an isotropic elastoplastic model. The main mechanical parameters are an elastic modulus E = 30 GPa, a Poisson's ratio ν = 0.22, and a density of 2500 kg / m³. 3 The maximum principal stress failure criterion based on fracture energy was adopted as the crack initiation criterion for the damaged material. The critical maximum principal stress was 2.34 MPa, the damage evolution type was energy, the softening type was linear softening, the fracture energy was 120 J, the viscosity coefficient was set to 1e-3, and the single crack model mesh adopted a C3D8R three-dimensional eight-node hexahedral mesh.

[0031] S23: Set the model load and set uniformly distributed compressive stress on the two sides of the concrete slab along the y direction. The direction of the uniformly distributed pressure is the center of the concrete slab, and the magnitude is the minimum compressive stress value of the whole ring determined in step S13. In this optional embodiment, a model load is set, including applying uniform compressive stress along both sides of the concrete slab towards the center of the concrete slab in the model, equivalent to the overall load effect of cracks causing a decrease in concrete strength, such as... Figure 5 As shown, a compressive stress is applied to the center of the surface of the concrete slab in the positive z-direction. The compressive stress value increases linearly with time until the concrete fails.

[0032] S24: Set the model boundary. Set the completely fixed boundary for the face of the four fixed concrete components that is in the opposite direction to the face that connects to the middle concrete slab. In this optional embodiment, boundaries are set at the concrete fixed ends along the x and y directions of the model, respectively; furthermore, as Figure 6 As shown, spring elements are used to simulate the constraint effect between a local area of ​​the tunnel lining and the surrounding structure; The spring stiffness is determined using the following formula: K = K 0×(0.197+0.803 e -20.426h / H ); In the formula, K 0 represents the original stiffness of the concrete structure, which is determined by the ratio of the concrete strength to the number of springs. h The depth of the crack. H The thickness of the concrete structure; The stiffness of the spring element is taken as the stiffness of the concrete or reinforced concrete when the crack penetrates; the spring elements are arranged on the four sides of the concrete slab opposite to the four concrete fixing members, one end of the spring element is connected to the concrete slab and the other end is connected to the concrete fixing members; the spring elements are used to constrain the concrete slab in three orthogonal directions.

[0033] S25: Set a single crack in the model, establish a precast crack at the center of the lower concrete slab surface based on the crack characteristics, and use the extended finite element module to set the precast crack as a crack, set the entire concrete slab as the extended finite element crack propagation area, and allow crack growth. In this optional embodiment, the impact of crack length on the safety of the lining structure is simulated by considering long cracks in segments; a precast crack is established at the center of the lower surface of the concrete slab in the negative z-axis direction, such as... Figure 7 As shown, the precast cracks were set as cracks using extended finite element method, and the entire concrete slab was set as the extended finite element crack propagation region, allowing crack growth.

[0034] S26: Using the static analysis in the ABAQUS analysis step module, apply compressive stress to the center of the surface of the concrete slab in the positive z-axis direction until the concrete fails, simulate the entire process of the propagation of a precast single crack in the concrete, and calculate and calculate the ultimate bearing capacity of the tunnel lining. In this optional embodiment, simulation calculations are performed using static analysis in the ABAQUS analysis step module to simulate the entire process of precast single crack propagation on concrete. A load is applied to the upper surface of the locally equivalent simplified model with and without cracks along the positive z-axis until the model fails. The ultimate bearing capacity of the locally equivalent simplified models with and without cracks is obtained respectively. Figure 8 As shown.

[0035] S27: The influence coefficient is calculated using the following formula: ; In the formula, α is the influence coefficient, F is the ultimate bearing capacity of the local equivalent simplified model of tunnel lining under single crack, and F0 is the ultimate bearing capacity of the local equivalent simplified model of tunnel lining without crack.

[0036] In an optional embodiment, considering the symmetry of the model, in S25, the crack angle is defined by the angle between the crack length direction and the longitudinal direction of the tunnel; the crack depth is defined by the maximum distance the crack extends from the lower surface of the concrete slab along the z-axis in a direction perpendicular to the surface into the interior of the concrete slab; and the crack direction is defined by the extension direction of the crack along the crack surface in terms of angle and depth. The crack angles are set to 0°, 10°, 20°, 30°, 40°, 50°, 60°, 70°, 80°, and 90°, respectively; and the crack depths are set to 0.05d, 0.1d, 0.2d, 0.3d, 0.5d, 0.7d, 0.8d, and 0.9d, respectively, where d is the original tunnel lining thickness. The three parameters are combined to form the input parameters for the simulation. Simulation calculations are performed in S26, and influence coefficients are calculated in S27.

[0037] In an optional embodiment, step S3 performs simulation calculations on the branch portion through the following steps: S31: Establish a locally equivalent simplified mathematical model of tunnel lining based on crack bifurcation morphology and crack depth. The dendritic crack morphology is represented by the combination of the angle between each crack in the branch and the tunnel circumferential direction. The locally equivalent simplified mathematical model is divided into a steel reinforcement model and a concrete model containing dendritic cracks. Define a Cartesian coordinate system and the model shape is a cuboid concrete block. In S3, for dendritic crack defects, the dendritic cracks are defined by branching parts and forked parts, and the impact of branching parts and forked parts on the safety status of tunnel lining is calculated separately; wherein, the branching part includes crack bifurcation points, and the forked part includes single cracks. The impact of the forked part on the safety status of tunnel lining can be determined according to the calculation method of the impact of single crack defects on the safety status of tunnel lining, specifically referring to S21-S27.

[0038] In this optional embodiment, the main body of the locally equivalent simplified mathematical model of the tunnel lining with branched sections is the concrete model and the reinforcement model with branched sections; based on the characteristics of dendritic crack tunnel lining, and using ABAQUS finite element simulation as a basis, global insertion is introduced. cohesive The unit simulates the crack morphology of the branched part, such as Figure 9 As shown, a locally equivalent simplified model is constructed to reflect the performance of the local structure when branches exist. Considering the reinforcement arrangement and the thickness of the protective layer, the three-dimensional parameters of the model are: length... l =1m, width w =0.45m, height t =1m.

[0039] S32: Set the model material parameters. The material parameters of the concrete model and the steel reinforcement model should be consistent with the tunnel lining material parameters. In this optional embodiment, the mechanical parameters of the concrete material are set using the properties module in ABAQUS, and the crack initiation criterion of the damaged material is defined. For the locally equivalent simplified model of dendritic cracks, the material properties are set as an elastoplastic model with surface forces. The main mechanical parameters are elastic stiffness Enn=30GPa, shear stiffness Ess=Ett=12.5GPa, and density of 2500 kg / m3. The maximum principal stress failure criterion based on fracture energy is adopted as the crack initiation criterion of the damaged material. The critical maximum principal stress is 2.5MPa, the maximum shear stress is 3.0MPa, the damage evolution type is energy, the softening type is linear softening, the fracture energy is 120J, and the viscosity coefficient is set to 1e-3. The dendritic crack model mesh adopts a C3D8R three-dimensional eight-node hexahedral mesh.

[0040] S33: Set the model boundary, setting a completely fixed boundary for all nodes on the lower surface of the model in the z-direction; In this optional embodiment, a Cartesian coordinate system is defined, and initial values ​​of concentrated force load and bending moment are set on the upper surface of the concrete in the positive z-axis direction in the model to calculate the ultimate bearing capacity of the model, restrict the displacement of the lower surface of the model in the negative z-axis direction in all directions, and prohibit its rotation.

[0041] S34: Set the dendritic cracks in the model; In this optional embodiment, dendritic cracks are provided in the model, such as Figure 10 As shown, cracks are prefabricated by deleting cohesive elements with a thickness of 0.

[0042] S35: Using the explicit dynamic analysis in the ABAQUS analysis step module, a combination of bending moment and axial force is applied to the center of the upper surface of the model along the positive z-axis until the model fails, simulating the entire process of the precast branch section's concrete expansion. First, a compressive load is applied to the upper surface of the locally equivalent simplified model of the branch section along the positive z-axis, resulting in point (0, N). Then, the ordinate of point (0, N) is divided into equal parts, and a bending moment is applied again with a fixed axial force value for one of the divisions until the locally equivalent simplified model of the branch section fails. The bending moment value at this point is recorded, and the combination of internal forces (M, N) at this point is taken as a point on the ultimate bearing capacity curve. Finally, the axial force is set to a fixed value of zero, and a pure bending moment is applied until the locally equivalent simplified model of the branch section fails. The bending moment value at this point is recorded, and the combination of internal forces (M, 0) at this point is the intersection of the ultimate bearing capacity curve and the horizontal axis, from which the ultimate bearing capacity curve of the substructure can be plotted. The influence coefficient can be determined by calculating the ratio of the area of ​​the ultimate bearing capacity curve under the condition of the locally equivalent simplified model without cracks to the area of ​​the ultimate bearing capacity curve under the condition of the locally equivalent simplified model with branch cracks. In this optional embodiment, when plotting the ultimate bearing capacity curve of the branch section in a two-dimensional coordinate system, a compressive load is first applied to the positive z-axis surface of the local equivalent simplified model of the branch section to obtain the point (0, N); then the ordinate of the point (0, N) is divided into equal parts, and a bending moment is applied with a certain axial force value as a constant until the local equivalent simplified model of the branch section fails. The bending moment value at this time is recorded, and the internal force combination (M, N) at this time is taken as a point on the ultimate bearing capacity curve; finally, the axial force is set to a constant value of zero, and a pure bending moment is applied until the local equivalent simplified model of the branch section fails. The bending moment value at this time is recorded, and the internal force combination (M, 0) at this time is the intersection of the ultimate bearing capacity curve and the abscissa, and the ultimate bearing capacity curve of the substructure can be plotted; by calculating the ratio of the area value of the ultimate bearing capacity curve with cracks to the area value of the ultimate bearing capacity curve without cracks, the influence coefficient can be determined.

[0043] S36: The influence coefficient is calculated using the following formula: ; In the formula, α is the influence coefficient, and S u S represents the area of ​​the ultimate bearing capacity curve of the locally equivalent simplified model of the tunnel lining with branched cracks, while S represents the area of ​​the ultimate bearing capacity curve of the locally equivalent simplified model of the tunnel lining without cracks.

[0044] In an optional embodiment, in S34, the crack angle is defined by the angle between the crack length direction and the tunnel longitudinal direction; the crack depth is defined by the maximum distance the crack extends from the lower surface of the concrete slab along the z-axis in a direction perpendicular to the surface into the interior of the concrete slab; and the crack direction is defined by the extension direction of the crack along the crack surface in terms of angle and depth. The crack angles are set to 10°, 20°, 30°, 40°, 60°, 70°, 80°, and 90°, respectively; and the crack depths are set to 0.05d, 0.1d, 0.3d, 0.5d, 0.7d, and 0.9d, respectively, where d is the original tunnel lining thickness. The three parameters are combined to form the input parameters for the simulation. Simulation calculations are performed in S35, and influence coefficients are calculated in S36.

[0045] In an optional embodiment, in S4, the neural network model is a BP neural network model with three layers: an input layer, a hidden layer, and an output layer. The input is the crack depth and crack angle, and the output is the influence coefficient.

[0046] A dataset was constructed using data obtained from S2 and S3 simulations. The dataset was divided into a training set and a test set. Different numbers of hidden layer nodes were set for training. During training, the connection weights and thresholds between neurons in each layer of the neural network model were adjusted to continuously reduce the error. Through repeated training and learning, the weights and thresholds were continuously adjusted to minimize the mean square error, so that the actual output of the BP neural network continuously approached the expected output. Finally, the number of hidden layer nodes was determined to be 6, at which point the mean square error of the training set was 0.0027028.

[0047] In an optional embodiment, in step S5, the remaining safety factor for each segment is calculated using the following formula: si = Si × α; In the formula, si is the residual safety factor, Si is the full-ring safety factor of the crack-free lining structure, and α is the influence factor.

[0048] In this optional embodiment, in view of the fact that the crack depth is inconsistent within the length range, the deepest position within the entire crack range is taken as the crack depth from a safety perspective; the long crack is divided into multiple single crack segments, and the remaining safety factor at the position of each single crack segment is determined, thereby obtaining the distribution of the remaining safety factor along the length direction of the entire long crack.

[0049] Based on crack characteristic parameters such as location, depth, orientation, and length, the residual safety factor distribution along the crack orientation is achieved. When the tunnel crack orientation is longitudinal, the crack's circumferential distribution position in the tunnel lining remains unchanged, and the residual safety factor along the crack length direction can be considered constant. When the tunnel crack orientation is oblique or circumferential, long cracks are subdivided and considered as composed of multiple subdivided units, such as... Figure 11As shown, each subdivided unit has a corresponding residual safety factor. Since each subdivided unit is located differently in the circumferential direction of the lining, the variation of the residual safety factor for each subdivided unit is determined by its corresponding distribution characteristics. When the crack direction deviates, it is only necessary to subdivide the crack and then calculate the residual safety factor for each subdivided unit according to the previous calculation method. Finally, the distribution of the residual safety factor of the tunnel crack is calculated, as shown below. Figure 12 As shown.

[0050] Based on the calculation results, the remaining safety factors were classified into four levels: Level I: 0 ≤ si < 2 is extremely dangerous, the tunnel lining structure is almost inoperable; Level II: 2 ≤ si < 6 is dangerous, the tunnel lining structure has a considerable probability of failure; Level III: 6 ≤ si < 10 is relatively safe, the tunnel lining structure functions normally and is unlikely to fail; Level IV: 10 ≤ si is safe, the tunnel lining structure functions normally. Figure 13 As shown, based on the remaining safety factor classification, the safety of the tunnel lining structure at different locations of cracks is classified to determine the service status of the tunnel lining structure.

[0051] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for assessing the safety status of tunnel lining with cracks, characterized in that, Includes the following steps: S1: Using the load-structure method and combining the actual design parameters of the tunnel to be evaluated, calculate the full-ring safety factor of the lining structure when there are no cracks. S2: For single-crack defects, a local equivalent simplified mathematical model and corresponding numerical simulation model of tunnel lining structure containing single cracks are established, and the influence coefficient of single crack on structural safety state under different crack characteristics is calculated. S3: For dendritic crack defects, the dendritic cracks are defined by branching parts and forked parts. For branching parts, a local equivalent simplified model of the dendritic crack is established, and the ultimate bearing capacity curve is plotted by applying different combinations of bending moment and axial force to the model. Then, the influence coefficient of the branching part of the dendritic crack is obtained. The calculation of the forked part uses the calculation method of single crack defects in step S2. A local equivalent simplified mathematical model and a corresponding numerical simulation model of the tunnel lining structure are established respectively. Then, the influence coefficient of the forked part of the dendritic crack is obtained. The influence coefficient of the forked part and the branching part on the structural safety state under different crack characteristics is calculated. S4: Construct and train a neural network model with crack geometry as input and influence coefficient as output, and use the data obtained from the simulation calculations in S2 and S3 to train the neural network model; S5: Divide the actual cracks into segments along the crack direction, extract the location of the cracks in different segments in the entire tunnel ring, calculate the influence coefficient of each segment based on the neural network model trained in S4, and calculate the remaining safety factor of each segment to obtain the distribution of the remaining safety factor.

2. The method for assessing the safety status of a cracked tunnel lining according to claim 1, characterized in that, S1 includes the following steps: S11: Establish a tunnel lining simulation model. Based on the characteristics of the tunnel lining and the actual situation, establish an ABAQUS finite element simulation model of the full-ring tunnel lining. S12: Set the surrounding rock pressure. Based on the calculation method of surrounding rock pressure in the railway tunnel design code, calculate the surrounding rock pressure for a general deep-buried tunnel to obtain design parameters such as vertical surrounding rock pressure and horizontal surrounding rock pressure. S13: Determine the minimum compressive stress value of the entire ring, extract the stress distribution through numerical simulation, and compare it with the railway tunnel design specifications to obtain the minimum stress value of the entire ring of the crack-free tunnel lining; S14: Calculate the safety factor without cracks. Based on the calculation method of structural safety factor in the railway tunnel design code, the safety factor of the above tunnel structure is calculated using the damage stage method.

3. The method for assessing the safety status of a tunnel lining with cracks according to claim 2, characterized in that, S2 includes the following steps: S21: The actual complete tunnel lining crack is equivalent to a local equivalent model. A three-dimensional reinforced concrete local equivalent simplified model of the single-crack tunnel lining structure is established. The reinforced concrete local equivalent simplified model includes a concrete model and a steel reinforcement model. A Cartesian coordinate system is defined. The model shape is: a concrete slab at the center, two concrete fixing members in the x and y directions of the concrete slab, and a total of four concrete fixing members. Each concrete fixing member is connected to the middle concrete slab by a spring. S22: Set the model material parameters. The material parameters of the concrete model and the steel reinforcement model should be consistent with the tunnel lining material parameters. S23: Set the model load and set uniformly distributed compressive stress on the two sides of the concrete slab along the y direction. The direction of the uniformly distributed pressure is the center of the concrete slab, and the magnitude is the minimum compressive stress value of the whole ring determined in step S13. S24: Set the model boundary. Set the completely fixed boundary for the face of the four fixed concrete components that is in the opposite direction to the face that connects to the middle concrete slab. S25: Set a single crack in the model, establish a precast crack at the center of the lower concrete slab surface based on the crack characteristics, and use the extended finite element module to set the precast crack as a crack, set the entire concrete slab as the extended finite element crack propagation area, and allow crack growth. S26: Using the static analysis in the ABAQUS analysis step module, apply compressive stress to the center of the surface of the concrete slab in the positive z-axis direction until the concrete fails, simulate the entire process of the propagation of a precast single crack in the concrete, and calculate the ultimate bearing capacity of the tunnel lining. S27: The influence coefficient is calculated using the following formula: ； In the formula, α is the influence coefficient, F is the ultimate bearing capacity of the local equivalent simplified model of tunnel lining under single crack, and F0 is the ultimate bearing capacity of the local equivalent simplified model of tunnel lining without crack.

4. The method for assessing the safety status of a cracked tunnel lining according to claim 3, characterized in that, In S24, spring units are used to simulate the constraint effect between the local area of ​​the tunnel lining and the surrounding structure; The spring stiffness is determined using the following formula: K = K 0×(0.197+0.803 e -20.426h / H )； In the formula, K 0 represents the original stiffness of the concrete structure, which is determined by the ratio of the concrete strength to the number of springs. h This represents the crack depth. H The thickness of the concrete structure; The stiffness of the spring element is taken as the stiffness of the concrete or reinforced concrete when the crack penetrates; the spring elements are arranged on the four sides of the concrete slab opposite to the four concrete fixing members, one end of the spring element is connected to the concrete slab and the other end is connected to the concrete fixing members; the spring elements are used to constrain the concrete slab in three orthogonal directions.

5. The method for assessing the safety status of a tunnel lining with cracks according to claim 3, characterized in that, In S25, the crack angle is defined as the angle between the crack length direction and the tunnel longitudinal direction; the crack depth is defined as the maximum distance the crack extends from the lower surface of the concrete slab along the z-axis in a direction perpendicular to the surface into the interior of the concrete slab; and the crack direction is defined as the direction of extension of the crack along the crack surface in terms of both angle and depth. The crack angles are set to 0°, 10°, 20°, 30°, 40°, 50°, 60°, 70°, 80°, and 90°, respectively; and the crack depths are set to 0.05d, 0.1d, 0.2d, 0.3d, 0.5d, 0.7d, 0.8d, and 0.9d, respectively, where d is the original tunnel lining thickness. These three parameters are combined to form the input parameters for the simulation. Simulation calculations are performed in S26, and influence coefficients are calculated in S27.

6. The method for assessing the safety status of a cracked tunnel lining according to claim 1, characterized in that, S3 performs simulation calculations on the branched portion through the following steps: S31: Establish a locally equivalent simplified mathematical model of tunnel lining based on crack bifurcation morphology and crack depth. The dendritic crack morphology is represented by the combination of the angle between each crack in the branch and the tunnel circumferential direction. The locally equivalent simplified mathematical model is divided into a steel reinforcement model and a concrete model containing dendritic cracks. Define a Cartesian coordinate system and the model shape is a cuboid concrete block. S32: Set the model material parameters. The material parameters of the concrete model and the steel reinforcement model should be consistent with the tunnel lining material parameters. S33: Set the model boundary, setting a completely fixed boundary for all nodes on the lower surface of the model in the z-direction; S34: Set the dendritic cracks in the model; S35: Using the explicit dynamic analysis in the ABAQUS analysis step module, a combination of bending moment and axial force is applied to the center of the upper surface of the model along the positive z-axis until the model fails, simulating the entire process of the precast branch section's concrete expansion. First, a compressive load is applied to the upper surface of the locally equivalent simplified model of the branch section along the positive z-axis, resulting in point (0, N). Then, the ordinate of point (0, N) is divided into equal parts, and a bending moment is applied again with a fixed axial force value for one of the divisions until the locally equivalent simplified model of the branch section fails. The bending moment value at this point is recorded, and the combination of internal forces (M, N) at this point is taken as a point on the ultimate bearing capacity curve. Finally, the axial force is set to a fixed value of zero, and a pure bending moment is applied until the locally equivalent simplified model of the branch section fails. The bending moment value at this point is recorded, and the combination of internal forces (M, 0) at this point is the intersection of the ultimate bearing capacity curve and the horizontal axis, from which the ultimate bearing capacity curve of the substructure can be plotted. The influence coefficient can be determined by calculating the ratio of the area of ​​the ultimate bearing capacity curve under the condition of the locally equivalent simplified model without cracks to the area of ​​the ultimate bearing capacity curve under the condition of the locally equivalent simplified model with branch cracks. S36: The influence coefficient is calculated using the following formula: ； In the formula, α is the influence coefficient, and S u S represents the area of ​​the ultimate bearing capacity curve of the locally equivalent simplified model of the tunnel lining with branched cracks, while S represents the area of ​​the ultimate bearing capacity curve of the locally equivalent simplified model of the tunnel lining without cracks.

7. The method for assessing the safety status of a cracked tunnel lining according to claim 6, characterized in that, In step S34, the crack angle is defined by the angle between the crack length direction and the tunnel longitudinal direction; the crack depth is defined by the maximum distance the crack extends from the lower surface of the concrete slab along the z-axis in a direction perpendicular to the surface into the interior of the concrete slab; and the crack direction is defined by the extension direction of the crack along the crack surface in terms of both angle and depth. The crack angles are set to 10°, 20°, 30°, 40°, 60°, 70°, 80°, and 90°, respectively; and the crack depths are set to 0.05d, 0.1d, 0.3d, 0.5d, 0.7d, and 0.9d, respectively, where d is the original tunnel lining thickness. These three parameters are combined to form the input parameters for the simulation. The simulation calculation is performed in step S35, and the influence coefficients are calculated in step S36.

8. The method for assessing the safety status of a cracked tunnel lining according to claim 1, characterized in that, The neural network model is a BP neural network model with three layers: an input layer, a hidden layer, and an output layer. The inputs are the crack depth and crack angle, and the output is the influence coefficient.

9. The method for assessing the safety status of a cracked tunnel lining according to claim 1, characterized in that, In step S5, the remaining safety factor for each segment is calculated using the following formula: s i =S i ×α; In the formula, s i S is the remaining safety factor. i α represents the safety factor for the crack-free lining structure across the entire ring, and α is the influence factor.

10. The method for assessing the safety status of a tunnel lining with cracks according to claim 9, characterized in that, In S5, the remaining safety factor is classified based on the calculation results: Level I: 0≤si<2 is very dangerous, the tunnel lining structure is almost ineffective; Level II: 2≤si<6 is dangerous, the tunnel lining structure has a considerable probability of failure; Level III: 6≤si<10 is relatively safe, the tunnel lining structure functions normally and is difficult to fail; Level IV: 10≤si is safe, the tunnel lining structure functions normally.

Images