Roadway surrounding rock mechanical parameter partition dynamic inversion method

Abstract

The invention relates to the technical field of civil and mining engineering, and discloses a roadway surrounding rock mechanical parameter zoning dynamic inversion method, which comprises the following steps: S1, acquiring roadway section shape and roadway surrounding rock grade data, determining a surrounding rock inversion boundary and an inversion moment according to the roadway surrounding rock grade and section shape, and carrying out an in-situ test in the inversion boundary to obtain an inversion result; acquiring ground stress test data and surrounding rock strain monitoring data at an inversion moment; and S2, drawing a strain-hole depth fitting curve according to the strain monitoring data at the inversion moment, determining the characteristic depth of the surrounding rock, dividing a surrounding rock inversion region, and establishing an inversion calculation model. According to the method, the current inversion region range and the softening damage coefficient are inverted, so that the surrounding rock mechanical parameters in each inversion region are determined, zoned dynamic acquisition of the tunnel surrounding rock mechanical parameters is realized, and surrounding rock mechanical parameter evolution of different regions of the tunnel surrounding rock can be timely and accurately mastered; and thus, dynamic evaluation of roadway stability and scientific decision-making of support can be realized.

Description

Technical Field

[0001] The invention relates to the technical field of civil and mining engineering, and in particular to a method for dynamic inversion of mechanical parameters of tunnel surrounding rocks by partitioning. Background Art

[0002] The surrounding rock structure of deep tunnel engineering is complex and has significant size effect. The parameters obtained through conventional rock mechanics tests are often quite different from the actual mechanical properties of the surrounding rock. Therefore, the surrounding rock mechanical parameter inversion method based on the tunnel surface displacement monitoring data has become a common means to obtain the mechanical properties of the surrounding rock of tunnel engineering. However, after the excavation of deep tunnel engineering, the surrounding rock stability evolution process gradually forms fracture and expansion zones, damage expansion zones and elastic zones, resulting in the surrounding rock mechanical parameters obtained by surface displacement inversion being an average and static mechanical parameter, which cannot truly reflect the dynamic surrounding rock mechanical properties of each zone during the evolution of surrounding rock stability. Therefore, it is necessary to invent a method for the dynamic inversion of tunnel surrounding rock mechanical parameters. This method can realize the dynamic acquisition of tunnel surrounding rock mechanical parameters in a zoned manner, providing a scientific basis for the long-term stability evaluation of tunnels and accurate support decision-making. Summary of the invention

[0003] In order to solve the technical problems existing in the prior art, the present invention provides a method for dynamic inversion of mechanical parameters of tunnel surrounding rock.

[0004] The present invention is implemented by the following technical solution: a method for dynamic inversion of mechanical parameters of tunnel surrounding rock, comprising the following steps: Step S1, collecting tunnel cross-sectional shape and tunnel surrounding rock grade data, determining the surrounding rock inversion boundary and inversion time according to the tunnel surrounding rock grade and cross-sectional shape, carrying out in-situ tests within the inversion boundary, and obtaining ground stress test data and surrounding rock strain monitoring data at the inversion time; Step S2, drawing a strain-hole depth fitting curve according to the strain monitoring data at the inversion time, determining the characteristic depth of the surrounding rock, dividing the surrounding rock inversion area, and establishing an inversion calculation model; Step S3: Based on the geostress data and the inversion calculation model, an orthogonal test method is used to simulate the tunnel excavation process and establish a surrounding rock mechanical parameter and surrounding rock strain sample database; Step S4, according to the strain monitoring data at the inversion time, the MEA-BP joint inversion method is used to determine the mechanical parameter values ​​of the surrounding rock in each inversion area in turn; Step S5, repeating steps S2 to S4, dynamically and regularly dividing the surrounding rock inversion area, and calculating the mechanical parameters in the inversion area until the strain monitoring is completed.

[0005] As a further improvement of the above scheme, the cross-sectional shape of the tunnel in step S1 is divided into straight wall semicircular arch, circle, and rectangle; the grade of the tunnel surrounding rock is divided into Class I, Class II, Class III, Class IV, and Class V.

[0006] As a further improvement of the above scheme, the inversion boundary in step S1 refers to the width, height, depth and center point of the numerical inversion calculation model; Among them, the width and height of the rock numerical inversion calculation model are consistent, the ratio of depth to lane width is 5-12:1; the ratio of width to lane width is 4-8:1; The width of the current lane is the diameter of the semicircle of a straight-wall semicircular arch lane, the diameter of the circle of a circular lane, and the length of the rectangle of a rectangular lane. The center point is the spatial position relationship between the centroid of the width and height section of the inversion calculation model and the tunnel section. The center point of the straight-wall semicircular arch tunnel coincides with the center of the semicircle, and the center point of the circular and rectangular tunnels coincides with the centroid of the tunnel section.

[0007] As a further improvement of the above scheme, the inversion time in step S1 refers to the time from the start of in-situ monitoring to the performance of numerical inversion calculation; The in-situ monitoring includes ground stress testing and surrounding rock strain monitoring. The surrounding rock strain monitoring adopts distributed optical fiber strain sensing technology or weak reflection optical fiber Bragg grating strain sensing technology, and the ground stress testing adopts the original rock stress testing technology of rheological stress recovery method.

[0008] As a further improvement of the above scheme, the characteristic depth of the surrounding rock in step S2 includes peak depth and yield depth, the peak depth refers to the hole depth corresponding to when the peak strain of the rock is equal to the strain value of the fitting curve, and the yield depth refers to the hole depth corresponding to when the yield strain of the rock is equal to the strain value of the fitting curve; the peak strain and yield strain are determined based on the total stress-strain curve of the rock under uniaxial compression conditions.

[0009] As a further improvement of the above solution, the inversion characteristic circle in step S2 includes a first characteristic circle, a second characteristic circle, a third characteristic circle and a fourth characteristic circle, all of which have the center point of the inversion calculation model as the center of the circle; Among them, the radius of the first characteristic circle is the distance from the center point to the maximum yield depth position, the radius of the second characteristic circle is the distance from the center point to the minimum yield depth position, the radius of the third characteristic circle is the distance from the center point to the maximum peak depth position, and the radius of the fourth characteristic circle is the distance from the center point to the minimum peak depth position; the maximum yield depth, minimum yield depth, maximum peak depth and minimum peak depth are the maximum and minimum values ​​of the yield depth and peak depth respectively.

[0010] As a further improvement of the above solution, the inversion region in step S2 includes an elastic inversion region, an elastic-damage inversion region, a damage inversion region, a damage-rupture inversion region and a rupture inversion region; Among them, the elastic inversion zone is the area between the first characteristic circle and the inversion boundary; the elastic-damage inversion zone is the area between the first characteristic circle and the second characteristic circle; the damage inversion zone is the area between the second characteristic circle and the third characteristic circle; the damage-fracture inversion zone is the area between the third characteristic circle and the fourth characteristic circle; and the fracture inversion zone is the area between the tunnel contour line and the fourth characteristic circle.

[0011] As a further improvement of the above scheme, in step S3, surrounding rock mechanical parameters are set, and the core material softening excavation simulation method is used to perform numerical simulation of tunnel excavation according to the softening range and softening damage coefficient of each inversion area of ​​the set surrounding rock mechanical parameters, so as to extract the surrounding rock hole depth-strain relationship results under different surrounding rock mechanical parameters to form a training sample set; the hole depth-strain relationship samples are used as the input set of the training sample set, and the surrounding rock mechanical parameters in each inversion area are used as the output set of the training sample set.

[0012] As a further improvement of the above solution, the step S4 of sequentially determining the mechanical parameter values ​​of the surrounding rock in each inversion area includes the following steps: Step A1, assuming that the elastic parameters of all inversion areas are the same, calculate the rock mechanical parameters in the elastic area based on the strain monitoring data in the elastic area; Step A2, based on the rock mechanical parameters in the elastic zone, assuming that the rock mechanical parameters in other inversion areas are the same, calculate the rock mechanical parameters in the elastic-damage zone based on the strain monitoring data in the elastic-damage zone; Step A3, based on the rock mass mechanical parameters in the elastic zone and the elastic-damaged zone, assuming that the rock mass mechanical parameters in other inversion areas are the same, the rock mass mechanical parameters in the damaged zone are calculated based on the strain monitoring data in the damaged zone; Step A4, based on the rock mechanical parameters of the elastic zone, the elastic-damage zone, and the damage zone, assuming that the rock mechanical parameters of other inversion areas are the same, calculate the rock mechanical parameters of the damage-rupture zone based on the strain monitoring data of the damage-rupture zone; Step A5: Based on the rock mechanical parameters of the elastic zone, elastic-damage zone, damage zone, and damage-fracture zone, assuming that the rock mechanical parameters of other inversion areas are the same, calculate the rock mechanical parameters of the fracture zone based on the strain monitoring data of the fracture zone.

[0013] As a further improvement of the above scheme, the step of dynamically and periodically dividing the surrounding rock inversion area in step S5 is to invert the range of each inversion area and the softening damage coefficient at the current moment according to the MEA-BP optimization inversion model trained in step S4 and the current surrounding rock hole depth-strain curve, so as to determine the surrounding rock mechanical parameters in each inversion area at the current moment; thereby realizing the dynamic inversion of the surrounding rock mechanical parameters.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention assigns the optimized initial weights and thresholds to the BP neural network model through the MEA algorithm, thereby avoiding the problem of local non-convergence or overfitting during the BP neural network inversion analysis, and establishes the nonlinear relationship between the mechanical parameters of the surrounding rock and the hole depth-strain curve through the MEA-BP optimization inversion model and sample set; The present invention uses the trained MEA-BP optimization inversion model and the current surrounding rock hole depth-strain curve to invert the current inversion area range and softening damage coefficient, thereby determining the surrounding rock mechanical parameters in each inversion area, and realizing the zoned dynamic acquisition of the tunnel surrounding rock mechanical parameters. It can timely and accurately grasp the evolution of the surrounding rock mechanical parameters in different areas of the tunnel surrounding rock, and further realize the dynamic evaluation of the tunnel stability and the scientific decision-making of support. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of a method for dynamic inversion of mechanical parameters of tunnel surrounding rock provided by the present invention; Figure 2 A schematic diagram of the boundary of the inversion calculation model in Example 2 provided by the present invention; Figure 3 A schematic diagram of the location of the center point of the lane in Example 3 provided by the present invention; Figure 4 A schematic diagram of the arrangement of monitoring drilling holes for a straight-wall semicircular arched tunnel in Example 3 provided by the present invention; Figure 5 A schematic diagram of the arrangement of circular tunnel monitoring drilling holes in Example 3 provided by the present invention; Figure 6 A schematic diagram of the arrangement of rectangular tunnel monitoring drilling holes in Example 3 provided by the present invention; Figure 7 A schematic diagram of the arrangement of the ground stress test section and the surrounding rock strain monitoring section in Example 3 provided by the present invention; Figure 8 A schematic diagram for determining the peak depth and yield depth in Example 4 provided by the present invention; Fig. 9 This is a schematic diagram of the inversion area division in Example 4 provided by the present invention. DETAILED DESCRIPTION

[0016] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form a new embodiment.

[0017] Example 1

[0018] Please combine Figure 1 A method for dynamic inversion of mechanical parameters of tunnel surrounding rock zoning in this embodiment includes the following steps: Step S1, collecting tunnel cross-sectional shape and tunnel surrounding rock grade data, determining the surrounding rock inversion boundary and inversion time according to the tunnel surrounding rock grade and cross-sectional shape, carrying out in-situ tests within the inversion boundary, and obtaining ground stress test data and surrounding rock strain monitoring data at the inversion time; The cross-sectional shapes of the tunnels are divided into straight wall semicircular arch, circle, and rectangle; the surrounding rock grades of the tunnels are divided into Class I, Class II, Class III, Class IV, and Class V; The inversion boundary refers to the width, height, depth and center point of the numerical inversion calculation model; Among them, the width and height of the rock numerical inversion calculation model are consistent, the ratio of depth to lane width is 5-12:1; the ratio of width to lane width is 4-8:1; The width of the current lane is the diameter of the semicircle of a straight-wall semicircular arch lane, the diameter of the circle of a circular lane, and the length of the rectangle of a rectangular lane. The center point is the spatial position relationship between the centroid of the width and height section of the inversion calculation model and the tunnel section. The center point of the straight-wall semicircular arch tunnel coincides with the center of the semicircle, and the center point of the circular and rectangular tunnels coincides with the centroid of the tunnel section. The inversion time refers to the time from the start of in-situ monitoring to the numerical inversion calculation; The in-situ monitoring includes ground stress testing and surrounding rock strain monitoring. The surrounding rock strain monitoring adopts distributed optical fiber strain sensing technology or weak reflection optical fiber Bragg grating strain sensing technology. The ground stress testing adopts the original rock stress testing technology of rheological stress recovery method. Step S2, drawing a strain-hole depth fitting curve according to the strain monitoring data at the inversion time, determining the characteristic depth of the surrounding rock, dividing the surrounding rock inversion area, and establishing an inversion calculation model; The characteristic depth of surrounding rock includes peak depth and yield depth. Peak depth refers to the hole depth corresponding to when the peak strain of rock is equal to the strain value of fitting curve, and yield depth refers to the hole depth corresponding to when the yield strain of rock is equal to the strain value of fitting curve. Peak strain and yield strain are determined according to the total stress-strain curve of rock under uniaxial compression conditions. The inversion characteristic circles include a first characteristic circle, a second characteristic circle, a third characteristic circle and a fourth characteristic circle, all of which have the center point of the inversion calculation model as the center of the circle; Among them, the radius of the first characteristic circle is the distance from the center point to the maximum yield depth position, the radius of the second characteristic circle is the distance from the center point to the minimum yield depth position, the radius of the third characteristic circle is the distance from the center point to the maximum peak depth position, and the radius of the fourth characteristic circle is the distance from the center point to the minimum peak depth position; the maximum yield depth, the minimum yield depth, the maximum peak depth and the minimum peak depth are the maximum and minimum values ​​of the yield depth and the peak depth respectively; The inversion area includes elastic inversion area, elastic-damage inversion area, damage inversion area, damage-rupture inversion area and rupture inversion area; Among them, the elastic inversion zone is the area between the first characteristic circle and the inversion boundary; the elastic-damage inversion zone is the area between the first characteristic circle and the second characteristic circle; the damage inversion zone is the area between the second characteristic circle and the third characteristic circle; the damage-rupture inversion zone is the area between the third characteristic circle and the fourth characteristic circle; the rupture inversion zone is the area between the roadway contour line and the fourth characteristic circle; Step S3: Based on the geostress data and the inversion calculation model, an orthogonal test method is used to simulate the tunnel excavation process and establish a surrounding rock mechanical parameter and surrounding rock strain sample database; Firstly, the surrounding rock mechanical parameters are set, and the core material softening excavation simulation method is used to perform numerical simulation of tunnel excavation according to the softening range and softening damage coefficient of each inversion area of ​​the set surrounding rock mechanical parameters, so as to extract the surrounding rock hole depth-strain relationship results under different surrounding rock mechanical parameters to form a training sample set; the hole depth-strain relationship samples are used as the input set of the training sample set, and the surrounding rock mechanical parameters in each inversion area are used as the output set of the training sample set; Step S4, according to the strain monitoring data at the inversion time, the MEA-BP joint inversion method is used to determine the mechanical parameter values ​​of the surrounding rock in each inversion area in turn; Determining the mechanical parameter values ​​of the surrounding rock in each inversion area in turn includes the following steps: Step A1, assuming that the elastic parameters of all inversion areas are the same, calculate the rock mechanical parameters in the elastic area based on the strain monitoring data in the elastic area; Step A2, based on the rock mechanical parameters in the elastic zone, assuming that the rock mechanical parameters in other inversion areas are the same, calculate the rock mechanical parameters in the elastic-damage zone based on the strain monitoring data in the elastic-damage zone; Step A3, based on the rock mass mechanical parameters in the elastic zone and the elastic-damaged zone, assuming that the rock mass mechanical parameters in other inversion areas are the same, the rock mass mechanical parameters in the damaged zone are calculated based on the strain monitoring data in the damaged zone; Step A4, based on the rock mechanical parameters of the elastic zone, the elastic-damage zone, and the damage zone, assuming that the rock mechanical parameters of other inversion areas are the same, calculate the rock mechanical parameters of the damage-rupture zone based on the strain monitoring data of the damage-rupture zone; Step A5, based on the rock mass mechanical parameters of the elastic zone, elastic-damage zone, damage zone, and damage-fracture zone, assuming that the rock mass mechanical parameters of other inversion areas are the same, calculate the rock mass mechanical parameters of the fracture zone based on the strain monitoring data of the fracture zone; Step S5, repeating steps S2 to S4, dynamically and regularly dividing the surrounding rock inversion area, and calculating the mechanical parameters in the inversion area until the strain monitoring is completed; The steps of dynamically and regularly dividing the surrounding rock inversion area are as follows: according to the MEA-BP optimized inversion model trained in step S4 and the current surrounding rock hole depth-strain curve, the range of each inversion area and the softening damage coefficient at the current moment are inverted, so as to determine the surrounding rock mechanical parameters in each inversion area at the current moment; and then the dynamic inversion of the surrounding rock mechanical parameters is realized.

[0019] Example 2

[0020] A method for dynamic inversion of tunnel surrounding rock mechanical parameters by zoning, comprising the following steps: Step S1: Collect the tunnel cross-sectional shape and tunnel surrounding rock grade data, and determine the surrounding rock inversion boundary (such as Figure 2 At the inversion time, in-situ tests are carried out within the inversion boundary to obtain ground stress test data and surrounding rock strain monitoring data at the inversion time; like Figure 2 As shown, B k is the width of the inversion calculation model; B s is the depth of the inversion calculation model; B g The height of the model for inversion calculation; Step S2, drawing a strain-hole depth fitting curve according to the strain monitoring data at the inversion time, determining the characteristic depth of the surrounding rock, dividing the surrounding rock inversion area, and establishing an inversion calculation model; Step S3: Based on the geostress data and the inversion calculation model, an orthogonal test method is used to simulate the tunnel excavation process and establish a surrounding rock mechanical parameter and surrounding rock strain sample database; Step S4, according to the strain monitoring data at the inversion time, the MEA-BP joint inversion method is used to determine the mechanical parameter values ​​of the surrounding rock in each inversion area in turn; Step S5, repeating steps S2 to S4, dynamically and regularly dividing the surrounding rock inversion area, and calculating the mechanical parameters in the inversion area until the strain monitoring is completed.

[0021] Example 3

[0022] The cross-sectional shapes of the tunnels are divided into straight wall semicircular arch, circle, and rectangle, and the surrounding rock grades of the tunnels are divided into Class I, Class II, Class III, Class IV, and Class V; The inversion boundary refers to the width, height, depth and center point of the numerical inversion calculation model; among them, the width of the numerical inversion calculation model of Class I, Class II, Class III, Class IV and Class V surrounding rock is 6 times, 7 times, 8 times, 9 times and 10 times the tunnel width respectively, the height is equal to the width, and the depth is 6 times the tunnel width; Lane width: for straight-walled semicircular arch lanes, it is the diameter of the semicircle; for circular lanes, it is the diameter of the circle; for rectangular lanes, it is the length of the rectangle; The center point is the spatial position relationship between the centroid of the width and height section of the inversion calculation model and the tunnel section. For straight-wall semicircular arch tunnels, the center point coincides with the center of the semicircle. For circular and rectangular tunnels, the center point coincides with the centroid of the tunnel section. Figure 3 As shown: SZ is a straight-wall semicircular arch section; SY is a circular section; SJ is a rectangular section; HK-Z is the width of a straight-wall semicircular arch tunnel; HK-Y is the width of a circular tunnel; HK-J is the width of a rectangular tunnel; ZX-Z is the center point of a straight-wall semicircular arch tunnel; ZX-Y is the center point of a circular tunnel; ZX-J is the center point of a rectangular tunnel.

[0023] The inversion time refers to the time from the start of in-situ monitoring to the numerical inversion calculation, including the first inversion time, the second inversion time, the third inversion time, etc.; among them, the first inversion time of Class I, II, and III surrounding rock does not exceed 10 days, and the interval between subsequent inversion times is 15 to 30 days; the first inversion time of Class IV and V surrounding rock does not exceed 5 days, and the interval between subsequent inversion times is 10 to 20 days.

[0024] The in-situ monitoring includes ground stress testing and surrounding rock strain monitoring. The ground stress testing and surrounding rock strain monitoring detection sections are distributed as follows: Figure 7 As shown, S-BX: surrounding rock strain monitoring section; S-YL: ground stress test section, in-situ monitoring includes three monitoring boreholes, namely the first monitoring borehole, the second monitoring borehole, and the third monitoring borehole, the borehole depths are 5 times, 4 times, and 4 times the tunnel width, respectively, the first monitoring borehole is a horizontal borehole, the second monitoring borehole is a vertical borehole, and the third monitoring borehole is 45±5° with the horizontal plane; among them, the first monitoring borehole, the second monitoring borehole, and the third monitoring borehole of the straight wall semicircular arch tunnel are arranged at the tunnel side, arch top, and shoulder, respectively, the first monitoring borehole, the second monitoring borehole, and the third monitoring borehole of the circular tunnel are arranged at 0°, 90°, and 135°, respectively, and the first monitoring borehole, the second monitoring borehole, and the third monitoring borehole of the rectangular tunnel are arranged at the tunnel side, the center of the roof, and the intersection of the side top, respectively; Three monitoring boreholes in the straight wall semicircular arch tunnel Figure 4 As shown, SZ-H: the first monitoring borehole of the straight-wall semicircular arch tunnel; SZ-V: the second monitoring borehole of the straight-wall semicircular arch tunnel; SZ-I: the third monitoring borehole of the straight-wall semicircular arch tunnel; The three monitoring boreholes in the circular tunnel are as follows: Figure 5 As shown, SY-H: the first monitoring borehole of the circular tunnel; SY-V: the second monitoring borehole of the circular tunnel; SY-I: the third monitoring borehole of the circular tunnel; SJ-H: The three monitoring boreholes in the rectangular tunnel are as follows: Figure 6As shown, the first monitoring borehole of the rectangular tunnel; SJ-V: the second monitoring borehole of the rectangular tunnel; SJ-I: the third monitoring borehole of the rectangular tunnel.

[0025] The surrounding rock strain monitoring adopts distributed optical fiber strain sensing technology or weak reflection optical fiber Bragg grating strain sensing technology, and the ground stress test adopts the original rock stress testing technology of rheological stress recovery method; among them, the ground stress test section is arranged within 1.5 meters of the bottom of the monitoring borehole, and the surrounding rock strain monitoring section is arranged from the mouth of the monitoring borehole to 1.5 meters from the bottom of the hole.

[0026] The geostress test data is the average value of the geostress test data of the first monitoring borehole, the second monitoring borehole, and the third monitoring borehole.

[0027] Example 4

[0028] The surrounding rock strain-hole depth fitting curve at the inversion time includes a first strain-hole depth fitting curve, a second strain-hole depth fitting curve, and a third strain-hole depth fitting curve, which are respectively obtained by fitting the surrounding rock strain monitoring data of the first monitoring borehole, the second monitoring borehole, and the third monitoring borehole; The surrounding rock strain-hole depth fitting curve at the inversion time is as follows Figure 8 As shown, εf is the peak strain value of rock; εq is the yield strain value of rock; D1 is the peak depth; D2 is the yield depth.

[0029] The characteristic depth of surrounding rock includes peak depth and yield depth; among them, peak depth refers to the hole depth corresponding to when the peak strain of rock is equal to the strain value of fitting curve, and yield depth refers to the hole depth corresponding to when the yield strain of rock is equal to the strain value of fitting curve; peak strain and yield strain are determined according to the total stress-strain curve of rock under uniaxial compression conditions.

[0030] The inversion characteristic circle includes the first characteristic circle, the second characteristic circle, the third characteristic circle and the fourth characteristic circle, all of which take the center point of the inversion calculation model as the center; wherein the radius of the first characteristic circle is the distance from the center point to the maximum yield depth position, the radius of the second characteristic circle is the distance from the center point to the minimum yield depth position, the radius of the third characteristic circle is the distance from the center point to the maximum peak depth position, and the radius of the fourth characteristic circle is the distance from the center point to the minimum peak depth position; the maximum yield depth, the minimum yield depth, the maximum peak depth and the minimum peak depth are the maximum and minimum values ​​of the yield depth and peak depth of the three monitoring boreholes respectively; The inversion area is divided into elastic inversion area, elastic-damage inversion area, damage inversion area, damage-fracture inversion area, and fracture inversion area; wherein the elastic inversion area refers to the area between the first characteristic circle and the inversion boundary, the elastic-damage inversion area refers to the area between the first characteristic circle and the second characteristic circle, the damage inversion area refers to the area between the second characteristic circle and the third characteristic circle, the damage-fracture inversion area refers to the area between the third characteristic circle and the fourth characteristic circle, and the fracture inversion area refers to the area between the roadway contour line and the fourth characteristic circle; The distribution of inversion characteristic circles and inversion areas is as follows: Fig. 9 As shown, Y1 is the first characteristic circle; Y2 is the second characteristic circle; Y3 is the third characteristic circle; Y4 is the fourth characteristic circle; Z1 is the elastic inversion zone; Z2 is the elastic-damage inversion zone; Z3 is the damage inversion zone; Z4 is the damage-rupture inversion zone; and Z5 is the rupture inversion zone.

[0031] Example 5

[0032] In the numerical simulation of deep tunnel excavation, the core material softening excavation simulation method is adopted, that is, the surrounding rock mechanical parameters in the tunnel excavation area are softened to 0, and the surrounding rock mechanical parameters are softened according to the range and damage coefficient of each inversion area shown in Table 1. That is, the numerical simulation of deep tunnel excavation is carried out based on the softening range and softening damage coefficient of each inversion area, where the softening range and softening damage coefficient variation range are shown in Table 2. Finally, according to Tables 1 and 2, the numerical simulation of deep tunnel excavation under different surrounding rock mechanical parameters based on the orthogonal test method is carried out, and the surrounding rock hole depth-strain relationship results under different surrounding rock mechanical parameters are extracted.

[0033] The sample database refers to a training sample set for inversion analysis based on the strain value results of different positions of the surrounding rock after tunnel excavation under different surrounding rock mechanical parameters. Among them, the hole depth-strain relationship sample is the input set of the training sample set, and the surrounding rock mechanical parameters in each inversion area are the output set.

[0034] Table 1 shows the range of surrounding rock mechanical parameters in each inversion area:

[0035] Table 2 is the numerical simulation test design table L(510) of surrounding rock mechanical parameters in each inversion area:

[0036] Example 6

[0037] The MEA-BP optimization inversion method is based on the BP neural network model. The optimized initial weights and thresholds are assigned to the BP neural network model through the MEA algorithm to avoid local non-convergence or over-fitting problems during the BP neural network inversion analysis. The nonlinear relationship between the mechanical parameters of the surrounding rock and the hole depth-strain curve is established through the MEA-BP optimization inversion model and sample set. Based on the trained MEA-BP optimization inversion model and the current surrounding rock hole depth-strain curve, the current inversion area range and softening damage coefficient are inverted to determine the mechanical parameters of the surrounding rock in each inversion area.

[0038] The mechanical parameter values ​​of the surrounding rock in each inversion area are determined in sequence, including the following steps: ① Assuming that the elastic parameters of all inversion areas are the same, the mechanical parameters of the rock mass in the elastic area are calculated according to the strain monitoring data of the elastic area; ② Based on the mechanical parameters of the rock mass in the elastic area, assuming that the mechanical parameters of the rock mass in other inversion areas are the same, the mechanical parameters of the rock mass in the elastic-damage area are calculated according to the strain monitoring data of the elastic-damage area; ③ Based on the mechanical parameters of the rock mass in the elastic area and the elastic-damage area, assuming that the mechanical parameters of the rock mass in other inversion areas are the same, the mechanical parameters of the rock mass in the damage area are calculated according to the strain monitoring data of the damage area; ④ Based on the mechanical parameters of the rock mass in the elastic area, the elastic-damage area, and the damage area, assuming that the mechanical parameters of the rock mass in other inversion areas are the same, the mechanical parameters of the rock mass in the damage-rupture area are calculated according to the strain monitoring data of the damage-rupture area; ⑤ Based on the mechanical parameters of the rock mass in the elastic area, the elastic-damage area, the damage area, and the damage-rupture area, assuming that the mechanical parameters of the rock mass in other inversion areas are the same, the mechanical parameters of the rock mass in the rupture area are calculated according to the strain monitoring data of the rupture area.

[0039] Example 7

[0040] Dynamically and regularly dividing the surrounding rock inversion area means that according to the trained MEA-BP optimization inversion model and the current surrounding rock hole depth-strain curve, the scope of each inversion area and the softening damage coefficient at the current moment are inverted, so as to determine the surrounding rock mechanical parameters in each inversion area at the current moment; thus realizing the dynamic inversion of surrounding rock mechanical parameters.

[0041] The present invention assigns the optimized initial weights and thresholds to the BP neural network model through the MEA algorithm, thereby avoiding the problem of local non-convergence or overfitting during the BP neural network inversion analysis, and establishes the nonlinear relationship between the mechanical parameters of the surrounding rock and the hole depth-strain curve through the MEA-BP optimization inversion model and sample set; Through the trained MEA-BP optimization inversion model and the current surrounding rock hole depth-strain curve, the current inversion area range and softening damage coefficient are inverted, so as to determine the surrounding rock mechanical parameters in each inversion area, realize the zoned dynamic acquisition of tunnel surrounding rock mechanical parameters, and timely and accurately grasp the evolution of surrounding rock mechanical parameters in different areas of tunnel surrounding rock, thereby realizing dynamic evaluation of tunnel stability and scientific decision-making on support.

[0042] The above-mentioned embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technicians in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. A method for dynamic inversion of tunnel surrounding rock mechanical parameters by zoning, characterized in that: The following steps are involved: Step S1, collecting tunnel cross-sectional shape and tunnel surrounding rock grade data, determining the surrounding rock inversion boundary and inversion time according to the tunnel surrounding rock grade and cross-sectional shape, carrying out in-situ tests within the inversion boundary, and obtaining ground stress test data and surrounding rock strain monitoring data at the inversion time; Step S2, drawing a strain-hole depth fitting curve according to the strain monitoring data at the inversion time, determining the characteristic depth of the surrounding rock, dividing the surrounding rock inversion area, and establishing an inversion calculation model; Step S3: Based on the geostress data and the inversion calculation model, an orthogonal test method is used to simulate the tunnel excavation process and establish a surrounding rock mechanical parameter and surrounding rock strain sample database; Step S4, according to the strain monitoring data at the inversion time, the MEA-BP joint inversion method is used to determine the mechanical parameter values ​​of the surrounding rock in each inversion area in turn; Step S5, repeating steps S2 to S4, dynamically and regularly dividing the surrounding rock inversion area, and calculating the mechanical parameters in the inversion area until the strain monitoring is completed.

2. A method for dynamic inversion of mechanical parameters of tunnel surrounding rock according to claim 1, characterized in that: In step S1, the cross-sectional shapes of the tunnel are divided into straight wall semicircular arch, circle, and rectangle; the grades of the tunnel surrounding rock are divided into Class I, Class II, Class III, Class IV, and Class V.

3. A method for dynamic inversion of tunnel surrounding rock mechanical parameters according to claim 1, characterized in that: The inversion boundary in step S1 refers to the width, height, depth and center point of the numerical inversion calculation model; Among them, the width and height of the rock numerical inversion calculation model are consistent, the ratio of depth to lane width is 5-12:1; the ratio of width to lane width is 4-8:1; The width of the current lane is the diameter of the semicircle of a straight-wall semicircular arch lane, the diameter of the circle of a circular lane, and the length of the rectangle of a rectangular lane. The center point is the spatial position relationship between the centroid of the width and height section of the inversion calculation model and the tunnel section. The center point of the straight-wall semicircular arch tunnel coincides with the center of the semicircle, and the center point of the circular and rectangular tunnels coincides with the centroid of the tunnel section.

4. A method for dynamic inversion of tunnel surrounding rock mechanical parameters according to claim 1, characterized in that: The inversion time in step S1 refers to the time from the start of in-situ monitoring to the performance of numerical inversion calculation; The in-situ monitoring includes ground stress testing and surrounding rock strain monitoring. The surrounding rock strain monitoring adopts distributed optical fiber strain sensing technology or weak reflection optical fiber Bragg grating strain sensing technology, and the ground stress testing adopts the original rock stress testing technology of rheological stress recovery method.

5. A method for dynamic inversion of tunnel surrounding rock mechanical parameters according to claim 1, characterized in that: The characteristic depth of the surrounding rock in step S2 includes the peak depth and the yield depth. The peak depth refers to the hole depth corresponding to when the peak strain of the rock is equal to the strain value of the fitting curve, and the yield depth refers to the hole depth corresponding to when the yield strain of the rock is equal to the strain value of the fitting curve; the peak strain and the yield strain are determined based on the total stress-strain curve of the rock under uniaxial compression conditions.

6. A method for dynamic inversion of mechanical parameters of tunnel surrounding rock according to claim 1, characterized in that: The inversion characteristic circles in step S2 include a first characteristic circle, a second characteristic circle, a third characteristic circle and a fourth characteristic circle, all of which have the center point of the inversion calculation model as the center of the circle; Among them, the radius of the first characteristic circle is the distance from the center point to the maximum yield depth position, the radius of the second characteristic circle is the distance from the center point to the minimum yield depth position, the radius of the third characteristic circle is the distance from the center point to the maximum peak depth position, and the radius of the fourth characteristic circle is the distance from the center point to the minimum peak depth position; the maximum yield depth, minimum yield depth, maximum peak depth and minimum peak depth are the maximum and minimum values ​​of the yield depth and peak depth respectively.

7. A method for dynamic inversion of mechanical parameters of tunnel surrounding rock according to claim 1, characterized in that: The inversion area in step S2 includes an elastic inversion area, an elastic-damage inversion area, a damage inversion area, a damage-rupture inversion area and a rupture inversion area; Among them, the elastic inversion zone is the area between the first characteristic circle and the inversion boundary; the elastic-damage inversion zone is the area between the first characteristic circle and the second characteristic circle; the damage inversion zone is the area between the second characteristic circle and the third characteristic circle; the damage-fracture inversion zone is the area between the third characteristic circle and the fourth characteristic circle; and the fracture inversion zone is the area between the tunnel contour line and the fourth characteristic circle.

8. A method for dynamic inversion of tunnel surrounding rock mechanical parameters according to claim 1, characterized in that: In the step S3, surrounding rock mechanical parameters are set, and the core material softening excavation simulation method is used to perform numerical simulation of tunnel excavation according to the softening range and softening damage coefficient of each inversion area of ​​the set surrounding rock mechanical parameters, so as to extract the surrounding rock hole depth-strain relationship results under different surrounding rock mechanical parameters to form a training sample set; the hole depth-strain relationship samples are used as the input set of the training sample set, and the surrounding rock mechanical parameters in each inversion area are used as the output set of the training sample set.

9. A method for dynamic inversion of mechanical parameters of tunnel surrounding rock according to claim 1, characterized in that: The step S4 of sequentially determining the mechanical parameter values ​​of the surrounding rock in each inversion area comprises the following steps: Step A1, assuming that the elastic parameters of all inversion areas are the same, calculate the rock mechanical parameters in the elastic area based on the strain monitoring data in the elastic area; Step A2, based on the rock mechanical parameters in the elastic zone, assuming that the rock mechanical parameters in other inversion areas are the same, calculate the rock mechanical parameters in the elastic-damage zone based on the strain monitoring data in the elastic-damage zone; Step A3, based on the rock mass mechanical parameters in the elastic zone and the elastic-damaged zone, assuming that the rock mass mechanical parameters in other inversion areas are the same, the rock mass mechanical parameters in the damaged zone are calculated based on the strain monitoring data in the damaged zone; Step A4, based on the rock mechanical parameters of the elastic zone, the elastic-damage zone, and the damage zone, assuming that the rock mechanical parameters of other inversion areas are the same, calculate the rock mechanical parameters of the damage-rupture zone based on the strain monitoring data of the damage-rupture zone; Step A5: Based on the rock mechanical parameters of the elastic zone, elastic-damage zone, damage zone, and damage-fracture zone, assuming that the rock mechanical parameters of other inversion areas are the same, calculate the rock mechanical parameters of the fracture zone based on the strain monitoring data of the fracture zone.

10. A method for dynamic inversion of mechanical parameters of tunnel surrounding rock according to claim 1, characterized in that: The step S5 of dynamically and regularly dividing the surrounding rock inversion area is to invert the range of each inversion area and the softening damage coefficient at the current moment according to the MEA-BP optimization inversion model trained in step S4 and the current surrounding rock hole depth-strain curve, so as to determine the surrounding rock mechanical parameters in each inversion area at the current moment; thereby realizing the dynamic inversion of the surrounding rock mechanical parameters.

Images