Tunnel supporting structure internal force regulation and control method and system

Abstract

The invention discloses a tunnel supporting structure internal force regulation and control method and system. The method comprises the following steps: constructing a plurality of sub-models based on the same neural network architecture to execute different functions respectively; in the primary support building process of the construction section, a target and an adjuster are installed; inputting the section deformation of the construction section into the first sub-model to obtain a continuous displacement field of the current construction section in the section direction; the second sub-model analyzes the continuous displacement field to obtain a continuous internal force field, and then the third sub-model generates a regulation and control scheme according to the continuous internal force field and predicts a preset position required for regulator installation in a next construction section. The system is used for executing the method. The beneficial effects of the invention are that the method can achieve the inversion of the full-field response of the initial support structure through the cooperation of a few data points with the mechanics principle, and can achieve the adaptive compensation of the internal force field of the tunnel surrounding rock deformation through the cooperation with corresponding hardware equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel construction, in particular to a tunnel supporting structure internal force regulation method and system. BACKGROUND

[0002] At present, when a tunnel is constructed, the initial supporting structure is generally a steel arch formed by splicing I-beams or a steel grid formed by splicing steel bar trusses, which is combined with sprayed concrete to bear the surrounding rock pressure. The disadvantage of the current technical form is that it can only passively bear the load of the surrounding rock, provides limited adjustment function, and has no deformation control function. However, in the actual construction process, due to different geological conditions, the deformation degree of different positions of the surrounding rock is different, which is easy to cause uneven stress of the initial supporting structure, causing uneven internal force of the overall structure, and leading to the occurrence of quality problems such as cracking, spalling and water leakage of the lining structure. SUMMARY

[0003] One of the purposes of the present application is to provide a tunnel supporting structure internal force regulation method which can solve at least one of the defects in the background art.

[0004] Another purpose of the present application is to provide a tunnel supporting structure internal force regulation system which can solve at least one of the defects in the background art.

[0005] In order to achieve at least one of the above purposes, the technical solution adopted by the present application is as follows: a tunnel supporting structure internal force regulation method, comprising the following steps: S100: According to the structural characteristics of the tunnel and the initial support, a plurality of sub-models based on the same neural network architecture are constructed, including a first sub-model for performing global inversion of displacement, a second sub-model for performing internal force analysis, and a third sub-model for performing adjustment response prediction; S200: Discretize the tunnel in space and time to obtain a plurality of continuous construction sections; during the construction process of the initial support of each construction section, install the adjustors required by the active adjustment network and the targets required by the displacement monitoring network to the predetermined positions; S300: During the construction process of the initial support of the entire construction section, continuously collect the cross-sectional deformation of the current construction section through the displacement monitoring network, and input the collected deformation data into the first sub-model to obtain the continuous displacement field of the current construction section; S400: The second sub-model analyzes the load distribution field obtained by the continuous displacement field inversion to obtain a continuous internal force field, and then the third sub-model generates a regulation scheme according to the continuous internal force field and predicts the predetermined position required for installing the adjustor of the next construction section; S500: The active adjustment network drives the adjustor installed at different predetermined positions of the current construction section according to the obtained regulation scheme.

[0006] Preferably, in step S100, each sub-model adopts a PINN architecture, and the specific construction includes the following processes: constructing a control equation according to the structural parameters of the primary support material and the structural parameters of the tunnel; giving the boundary of the control equation according to the construction method of the tunnel; obtaining a PINN basic model based on the combination of the DNN basic network model and the control equation and the boundary; setting multiple loss functions to optimize the PINN basic model, and obtaining the required first sub-model, second sub-model and third sub-model based on the required loss function types and / or weights.

[0007] Preferably, the loss function includes a physical equation residual loss, a boundary condition residual loss, an initial condition residual loss and a real acquisition data residual loss; wherein the PINN basic model is optimized by the physical equation residual loss, the boundary condition residual loss and the real acquisition data residual loss to obtain the first sub-model; the PINN basic model is optimized by the physical equation residual loss and the boundary condition residual loss, and the second sub-model and the third sub-model are obtained according to the different weights of the physical equation residual loss and the boundary condition residual loss.

[0008] Preferably, in the total loss function L tol of the first sub-model, the weights of the physical equation residual loss L phy and the real acquisition data residual loss L true are adaptive weights, and the specific obtaining process is as follows: a trainable noise parameter is automatically learned for each loss function by using homoscedastic uncertainty, and the expression of L tol of the improved total loss function of the first sub-model is as follows: ; Wherein, σ phy and σ true represent the noise parameters corresponding to the physical equation residual loss and the real acquisition data residual loss respectively, L bc represents the boundary condition residual loss, and λ bc represents the weight of the boundary condition residual loss.

[0009] Preferably, when installing the target, multiple point positions are set as the predetermined positions of the target installation in the cross-sectional direction of the current construction section of the tunnel at equal intervals; wherein the midpoint position of the cross section is one of the point positions, and the remaining point positions are symmetrically distributed to both sides of the midpoint of the cross section.

[0010] Preferably, the first sub-model obtains the deformation data of the predetermined positions collected by the displacement monitoring network, and then obtains the continuous displacement field by interpolating the deformation of adjacent predetermined positions along the cross-sectional direction of the construction section; when the construction length of the tunnel is less than a set value of the total length of the tunnel, the initial PINN base model is trained by the measured data in the initial stage of tunnel construction; the trained initial PINN base model is used as a physical discriminator and connected in parallel with a traditional discriminator to construct a generative adversarial network discriminator; the control equation of each sub-model is embedded into the generative adversarial network discriminator, and then the overfitting is suppressed by the generative adversarial network discriminator during the construction of the continuous displacement field by the first sub-model.

[0011] Preferably, a plurality of installation stations for installing the adjuster are arranged at equal intervals on the arch frame used for the initial support construction of the construction section; the third sub-model outputs the internal force peak point of the continuous internal force field according to the continuous internal force field output by the second sub-model; and the installation station closest to the internal force peak point is selected as the installation predetermined position of the adjuster of the next construction section.

[0012] A tunnel support structure internal force regulation system for implementing the tunnel support structure internal force regulation method described above, comprising a displacement monitoring network, an active regulation network, and an edge analysis module; the displacement monitoring network is used to collect the cross-sectional deformation of the current construction section of the tunnel; the active regulation network compensates the continuous internal force field of the current construction section according to the received internal force regulation signal; the edge analysis module is respectively connected with the displacement monitoring network and the active regulation network, and is used to construct a plurality of sub-models based on the same neural network architecture; the edge analysis module is adapted to receive the collected data of the displacement monitoring network, and then generate the internal force regulation signal according to the constructed sub-models and send it to the active regulation network.

[0013] Preferably, the displacement monitoring network comprises a displacement collection component installed on a drainage plate trolley and a target installed on the tunnel construction section; the displacement collection component is adapted to move the drainage plate trolley, and then identify the target installed on the current construction section.

[0014] Preferably, the active regulation network comprises a hydraulic station installed on the drainage plate trolley, and a regulator installed on the arch required for the construction of the primary support; the regulator comprises a jacking assembly and a hydraulic loading assembly; the jacking assembly is detachably installed on the arch, and is adapted to extend within a set adjustment range and mechanically locked when reaching a set extension amount; the hydraulic loading assembly is detachably installed on the driving end of the jacking assembly to drive the jacking assembly to extend; the hydraulic loading assembly is adapted to be detached from the jacking assembly after the jacking assembly is mechanically locked; the hydraulic station is adapted to receive the internal force regulation signal of the edge analysis module, and is adapted to be connected with the hydraulic loading assembly of the regulator in oil circuit, so as to control the loading amount of the hydraulic loading assembly according to the internal force regulation signal.

[0015] Compared with the prior art, the application has the beneficial effects that: Compared with the traditional method, the application can obtain the full-field response of the primary support structure by fewer data points and mechanical principles, and then cooperate with the corresponding hardware equipment to adaptively compensate the internal force field of the tunnel surrounding rock deformation, so as to effectively save the equipment cost while ensuring the safety of the surrounding rock primary support structure; and the effect of compensating while constructing the tunnel can also be realized. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 It is a working step schematic diagram of the regulation method of the application.

[0017] Figure 2 It is a PINN architecture schematic diagram of each sub-model in the application.

[0018] Figure 3 It is a simple structure schematic diagram of the tunnel construction in the application.

[0019] Figure 4 It is a simple structure schematic diagram of the tunnel section direction in the application.

[0020] Figure 5 It is a structure schematic diagram of the regulator in the application.

[0021] In the figure: surrounding rock 01, target 011, arch 02, installation station 021, drainage plate trolley 03, displacement acquisition assembly 04, hydraulic station 05, edge analysis module 06, support 1, base 11, outer sleeve 12, jacking piece 2, locking piece 3, hydraulic loading assembly 4, hydraulic cylinder 41, connecting assembly 42. DETAILED DESCRIPTION

[0022] Hereinafter, the present application will be further described with reference to the specific embodiments, it should be noted that in the description of the present specification, the description referring to the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in the present specification.

[0023] In the description of the present application, it should be noted that for orientation words such as the terms "center", "transverse", "longitudinal", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise" and the like indicate the orientation and positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and cannot be understood as limiting the specific protection scope of the present application.

[0024] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence.

[0025] In the present application, unless otherwise specified and limited, the terms "mounting", "connection", "connection", "fixing" and the like should be understood in a broad sense, for example, it can be connected, or detachable, or integrated; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, it can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0026] In the present application, unless specifically stated and limited otherwise, the first feature is "on" or "under" the second feature can include that the first and second features are in direct contact, or can include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, the first feature is "on", "above" and "over" the second feature includes that the first feature is directly above and obliquely above the second feature, or only indicates that the first feature is higher in horizontal height than the second feature. The first feature is "under", "below" and "underneath" the second feature includes that the first feature is directly below and obliquely below the second feature, or only indicates that the first feature is lower in horizontal height than the second feature.

[0027] The terms "comprising" and "having" and any variations thereof in the specification and claims of the present application are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or apparatus that includes a list of steps or units not necessarily limited to those clearly listed, but can include other steps or units not clearly listed or inherent to such processes, methods, products or apparatuses.

[0028] One aspect of the present application provides a method for regulating internal forces of a tunnel support structure, as shown in Figure 1 and Figure 4 A preferred embodiment includes the following steps: S100: According to the structural characteristics of the tunnel and the primary support, a plurality of sub-models based on the same neural network architecture are constructed, including a first sub-model for performing global inversion of displacement, a second sub-model for performing internal force analysis, and a third sub-model for performing adjustment response prediction.

[0029] S200: The tunnel is discretized in space and time to obtain a plurality of continuous construction sections; during the construction of the primary support in each construction section, the adjustors required by the active adjustment network and the targets required by the displacement monitoring network are installed to the predetermined positions.

[0030] S300: During the construction of the primary support in the entire construction section, the cross-sectional deformation of the current construction section is continuously collected by the displacement monitoring network, and the collected deformation data is input into the first sub-model to obtain the continuous displacement field of the current construction section.

[0031] S400: The second sub-model analyzes the load distribution field obtained by the continuous displacement field inversion to obtain the continuous internal force field, and then the third sub-model generates a regulation scheme according to the continuous internal force field and predicts the predetermined position required for the installation of the adjustor in the next construction section.

[0032] S500: The active adjustment network drives the adjustors installed at different predetermined positions in the current construction section according to the obtained regulation scheme.

[0033] It can be understood that the analysis method for the deformation of the tunnel surrounding rock 01 in the conventional method mainly adopts a pure mechanical method and a pure data-driven method. For the pure mechanical method, the numerical analysis of the internal force of the tunnel is mainly performed through the finite element software; however, this method needs manual mesh partitioning and boundary setting operations, and it is difficult to timely guide the adjustment of the internal force in the compact tunnel construction process. For the pure data-driven method, the data points required for the deep learning of the tunnel internal force are used for model prediction; however, the pure data-driven method requires a large number of data points, and sometimes needs to measure the inside of the surrounding rock 01, in addition, the response of the surrounding rock 01 is generally lagging behind the initial support construction process, so it is impossible to guide the internal force adjustment of the initial support structure.

[0034] In the technical solution of the present application, the data-driven model based on the physical equation, i.e., the first sub-model to the third sub-model, can be constructed through the structural characteristics of the tunnel and the initial support, and then the target 011 installed at the predetermined position of the tunnel section in the initial stage of the initial support construction of the construction section, so as to realize the collection of the deformation data of the partial position of the tunnel section. According to the less data points obtained by the collection, the full-field response of the initial support structure can be inversely obtained through the less data points and the mechanical principle, and then cooperating with the corresponding hardware equipment, the internal force field of the tunnel surrounding rock 01 deformation can be adaptively compensated. Moreover, according to the full-field response of the current initial support structure, the installation number and position of the hardware equipment of the next construction section can be predicted in advance, so as to effectively save the equipment cost under the condition of ensuring the safety of the surrounding rock 01 initial support structure. At the same time, the effect of compensating while constructing the tunnel can also be realized. In order to facilitate understanding, the steps of the technical solution of the present application will be described in detail below.

[0035] In the present embodiment, there are various types of neural network architectures for realizing the corresponding functions of each sub-model, and the PINN architecture, i.e., the physical information neural network architecture, is preferably adopted, which belongs to the extension of the deep neural network (DNN), i.e., by adding physical rules or constraints to the basic network model of the DNN, the required PINN basic model can be obtained. The construction of each sub-model in the present embodiment includes the following processes: constructing the control equation according to the structural parameters of the initial support material and the structural parameters of the tunnel. The boundary of the control equation is given according to the construction method of the tunnel. Based on the combination of the DNN basic network model and the control equation and the boundary, the PINN basic model is obtained. A plurality of loss functions are set to optimize the PINN basic model, and based on the required loss function types and / or weights for optimization, the required first sub-model, second sub-model and third sub-model are obtained.

[0036] It can be understood that, since the construction of the tunnel is carried out in sections, and there may be certain differences in the physical parameters between the sections of the tunnel, when the PINN basic model is constructed, it can be continuously carried out based on the tunnel construction process; that is, the PINN basic model is constructed through the structural parameters of the initial construction section of the tunnel, and as the tunnel construction continues, the already constructed PINN basic model is continuously updated through the structural parameters of the subsequent construction section of the tunnel, to ensure that the PINN basic model always corresponds to the current construction section parameters.

[0037] Therefore, before the construction of the PINN basic model, the tunnel needs to be discretized in space-time domain based on the construction requirements of the tunnel. Specifically, a suitable coordinate is selected, and uniform sampling is performed in the length domain of the tunnel to realize the spatial domain discretization of the tunnel; then each sampling point is discretized in time domain according to the on-site construction process, and the space-time coordinates of each sampling point are obtained. It should be noted that the tunnel section corresponding to adjacent sampling points is a single construction section; the length of the single construction section can be selected by the person skilled in the art according to actual needs, and according to industry experience, the length of the single construction section can be 10m.

[0038] In this embodiment, for the construction of the sub-model, taking an arbitrary construction section as an example, first, the required parameters need to be obtained, including the equivalent elastic modulus E of the required primary support material, the cross-sectional thickness A of the primary support, the cross-sectional moment of inertia I, the primary support calculation width B, the foundation modulus K of the tunnel, the cross-sectional radius p of the tunnel, the central angle θ corresponding to the cross section of the tunnel, the tangential uniform load distribution function q ρ and q θ Based on the above parameters, the control equation expression is as follows: ; .

[0039] Where Q represents shear force, which is derived from the bending moment M, the bending moment , k represents curvature; N represents axial force, , and ε represents tangential strain.

[0040] ; .

[0041] Where u represents tangential displacement, and w represents radial displacement.

[0042] It can be understood that, after obtaining the control equation, the boundary control equation can be determined according to the actual working condition; for the convenience of understanding, the following will take the construction stage of the drill-and-blast tunnel as an example, without considering the optimal locking anchor rod and the shear stiffness of the lining-surrounding rock 01 interface, then the boundary conditions are: u=0, M=0, Q=0; if the tunnel cross section is a semicircle, the boundary range of the central angle θ is [0, π].

[0043] Based on the above control equation and boundary conditions, a DNN basic network model as shown in FIG. 1 can be constructed. The DNN basic network model includes an input layer x, a hidden layer and an output layer y; wherein the input layer x of the DNN basic network model is the coordinates of the sampling points of the deformation of the cross section of the construction section, x = [p, q] T , and the output layer y is the bidirectional displacement of all sampling points, y = [w, u] T . After combining the DNN basic network model with the control equation and the boundary conditions, a PINN basic model of an elastic foundation beam can be formed. It should be noted that the loss function corresponding to the PINN basic model includes a physical equation residual loss L phy , a boundary condition residual loss L bc , an initial condition residual loss L ini and a real data residual loss L true ; the specific expressions of the loss functions are known to those skilled in the art, and therefore will not be described in detail here. According to the types and weight combinations of the four loss functions given, different total loss functions L tol can be obtained. The total loss function L tol obtained is back-propagated to the DNN basic network model, and the first sub-model D-PINN, the second sub-model F-PINN and the third sub-model A-PINN required can be obtained.

[0044] Specifically, the PINN basic model is optimized by the physical equation residual loss, the boundary condition residual loss and the real data residual loss, and the first sub-model D-PINN performing global inversion of displacement is obtained. That is, the total loss function L corresponding to the first sub-model D-PINN; wherein l phy , l bc , l true respectively represent the weights corresponding to the physical equation residual loss L phy , the boundary condition residual loss L bc and the real data residual loss L true .

[0045] The PINN basic model is optimized by the physical equation residual loss and the boundary condition residual loss, and according to the different weights of the physical equation residual loss and the boundary condition residual loss, the second sub-model F-PINN performing internal force analysis and the third sub-model A-PINN performing adjustment response prediction are obtained. That is, the expressions of the total loss functions L tol corresponding to the second sub-model F-PINN and the third sub-model A-PINN are both: ; only the weights l phy are different for different sub-models.phy and λ bc The values ​​can be different.

[0046] To facilitate understanding, the following is a detailed description of the specific working process of the first sub-model D-PINN, the second sub-model F-PINN, and the third sub-model A-PINN after their construction is completed.

[0047] like Figure 3 As shown, the construction section of the tunnel is represented by D. As the tunnel construction progresses, the construction section continues from D1 to D2. i That is, the current construction section is D. i At this point, the input parameter for the first sub-model D-PINN is the equivalent elastic modulus E of the initial support material. i initial support section thickness A i Moment of inertia I of cross section i Initial support calculation width B i (Generally taken as 1m), foundation modulus K i Monitoring point coordinates x ture and the corresponding displacement data w ture u ture The output of the first sub-model D-PINN is the construction section D. i The uniformly distributed load distribution function q of the cross section ρ (θ ij ), q θ (θ ij ), displacement distribution function w(θ) ij ), u(θ) ij ), θ ij Indicates construction section D i The central angles on the cross section corresponding to different target 011 installation positions; where, the uniformly distributed load distribution function q on the cross section. ρ (θ ij ), q θ (θ ij The displacement distribution function w(θ) ij ), u(θ) ij The result is obtained through inversion.

[0048] The input parameter for the second sub-model F-PINN is the equivalent elastic modulus E of the initial support material. i initial support section thickness A i Moment of inertia of cross section I i Initial support calculation width B i (Generally taken as 1m), foundation modulus K i Construction section D i The uniformly distributed load distribution function q of the cross section ρ (θ ij ), q θ (θ ij). The output of the second sub-model F-PINN is the cross-section internal force distribution function N(θ i ), M(θ ij ), Q(θ ij ) of the construction section D ij .

[0049] The input parameters of the third sub-model A-PINN are the equivalent elastic modulus E i of the primary support material, the cross-section thickness A i of the primary support, the cross-section moment of inertia I i , the primary support calculation width B i (typically 1m), the foundation modulus K i , the cross-section internal force distribution function N(θ i ), M(θ ij ), Q(θ ij ) of the construction section D ij , and the displacement distribution function w(θ ij ), u(θ ij ). The output of the third sub-model A-PINN is the adjustment load size of the installed adjuster at each predetermined position, and the predicted installation position of the adjuster corresponding to the next construction section D i+1 .

[0050] In this embodiment, as can be known from the working process of each sub-model described above, the first sub-model D-PINN is the first ring of the entire working process, and the accuracy of the output results of the first sub-model D-PINN directly affects the result accuracy of the subsequent second sub-model F-PINN and the third sub-model A-PINN. Since the tunnel is still in the construction state during the data collection process of the first sub-model D-PINN, based on the possible data jump points that may occur during the construction process, such as when using blasting construction, the vibration generated by blasting will cause the displacement data of the cross-section monitoring points of the current construction section to mutate, and this mutation result will cause the real collected data residual loss L true corresponding weight λ true to be contaminated, which forces the first sub-model D-PINN to fit the wrong data, thereby reducing the output accuracy of the first sub-model D-PINN. Therefore, when constructing the total loss function L tol corresponding to the first sub-model D-PINN, the weights of the physical equation residual loss L phy and the real collected data residual loss L true are adaptive weights, which can adaptively change the weight coefficient when the monitoring point data mutates, to ensure the accuracy of the output results.

[0051] Specifically, for the physical equation residual loss L phy and the real collected data residual loss L trueThe adaptive weight acquisition process is: using homoscedastic uncertainty to respectively calculate the physical equation residual loss L phy and the real acquisition data residual loss L true An automatic learning of a trainable noise parameter σ phy and σ true is performed, so that each residual loss term automatically reduces the confidence of high-noise data, and the L tol of the improved total loss function of the first sub-model D-PINN is expressed as: .

[0052] It can be understood that the noise parameter σ phy and σ true can be updated together with other weight coefficients through back propagation, without manual parameter tuning. Based on the improved total loss function of the first sub-model D-PINN, when the real acquisition data of a batch of monitoring points suddenly changes, the corresponding noise parameter σ true will increase, thereby reducing the corresponding weight coefficient of the real acquisition data residual loss L true , and automatically reducing the influence of the observation value on the output accuracy of the first sub-model D-PINN. If the physical equation residual continues to increase, the corresponding noise parameter σ phy will also continue to increase, so that the corresponding weight coefficient of the physical equation residual loss L phy is reduced.

[0053] In the embodiment, the monitoring points required by the first sub-model D-PINN are the points of the target 011 installed on the surrounding rock 01 during the initial support construction process of the entire construction section, so the position of the monitoring point is the displacement of the target 011, and it is also the deformation amount of the surrounding rock 01 corresponding to the installation point of the target 011. Therefore, the target 011 needs to be installed on the surrounding rock 01 in the initial stage of the initial support construction to ensure that the total deformation amount of the surrounding rock 01 during the entire initial support construction process can be monitored. Figure 4 As shown in the figure, when the target 011 is installed, multiple points are arranged as the installation predetermined positions of the target 011 at equal intervals in the cross-sectional direction of the current construction section of the tunnel; wherein the midpoint position of the cross section is one of the point positions, and the remaining point positions are symmetrically distributed to both sides of the midpoint of the cross section.

[0054] It should be known that the specific number and position of the target 011 can be set by the actual needs of those skilled in the art, the more the number of targets 011, the more the real data acquisition of the corresponding monitoring point, and the more accurate the output result of the first sub-model D-PINN, but too many targets 011 will cause data redundancy. Therefore, in the embodiment, the number of targets 011 is preferably five, and the midpoint of the cross section of the construction section corresponds to one installation point, and the other four points are divided into two groups, and are respectively arranged on both sides of the highest point of the cross section. With the midpoint of the cross section of the construction section extending to both sides, the central angle of the installation point of the adjacent target 011 is 20°-30°.

[0055] It can be understood that after the first sub-model D-PINN obtains the deformation data of the predetermined position collected by the displacement monitoring network, the deformation data of the adjacent predetermined position is interpolated along the cross section direction of the construction section, and then the continuous displacement field required, i.e. the displacement distribution function w(θ ij ), u(θ ij ) is obtained. However, in the initial stage of tunnel construction, especially in the process of building the first initial support, the displacement data of the target 011 is obtained for the first time by the first sub-model D-PINN, and the interpolation of the continuous displacement field based on the displacement data is prone to overfitting; as the number of construction sections continues to increase, the first sub-model D-PINN can update the fitting process of the continuous displacement field iteratively according to the continuously obtained displacement data of the target 011 of different construction sections, so that the fitted continuous displacement field meets the accuracy requirements. Therefore, in the initial stage of tunnel construction, i.e. when the construction length of the tunnel is less than the set value of the total length of the tunnel, such as when the construction length of the tunnel is less than 20% of the total length of the tunnel, the fitting process of the first sub-model D-PINN to obtain the continuous displacement field needs to be modified to ensure the output accuracy of the first sub-model D-PINN.

[0056] In the embodiment, the fitting modification method of the first sub-model D-PINN to obtain the continuous displacement field in the initial stage of tunnel construction is as follows: training an initial PINN base model through the measured data in the initial stage of tunnel construction; and connecting the trained initial PINN base model as a physical discriminator with a traditional discriminator to construct a generative adversarial network discriminator. The control equation for constructing each sub-model is embedded into the generative adversarial network discriminator, and then the overfitting is inhibited by the generative adversarial network discriminator in the process of constructing the continuous displacement field by the first sub-model D-PINN. For the convenience of understanding, the specific parameters will be described in detail below.

[0057] Specifically, the first construction section, i.e., the tunnel construction 10m, is used to train the initial PINN base model with the measured data, and the initial PINN base model is used as a physical discriminator in parallel with the traditional discriminator to obtain a generative adversarial network discriminator. Every 5m of tunnel construction, the generative adversarial network discriminator generates a plurality of groups of pseudo displacement sample data for fitting the displacement continuous field. At the same time, the pseudo displacement sample data obtained is mixed with the real data collected in a 1:1 ratio as a training set for training and updating the generative adversarial network discriminator.

[0058] In the embodiment, as shown in FIG. 2, a plurality of installation stations 021 for installing the adjuster are arranged at equal intervals on the arch frame 02 used for the construction of the primary support; the peak point of the internal force field of the continuous internal force field output by the second sub-model F-PINN is output by the third sub-model A-PINN; and the installation station 021 closest to the peak point of the internal force is selected as the installation predetermined position of the adjuster of the next construction section. Figure 4

[0059] It can be understood that, in the process of primary support construction, the arch frame 02 is used for supporting the entire surrounding rock 01, and the gap between the arch frame 02 and the surrounding rock 01 is small due to the fitting support between the arch frame 02 and the surrounding rock 01 through the adjuster. Therefore, the installation of the adjuster is usually performed before the arch frame 02 is erected, i.e., the adjuster is installed on the arch frame 02, and then the arch frame 02 with the installed adjuster is used for internal force field compensation of the surrounding rock 01 corresponding to the current construction section. In order to improve the construction efficiency of the tunnel, the installation of the adjuster is performed during the construction of the primary support, and the timing of the output of the continuous displacement field by the first sub-model D-PINN is when the construction of the primary support is just completed, i.e., the installation of the adjuster is ahead of the output of the first sub-model D-PINN, so the installation position of the adjuster needs to be predicted in advance. In the entire tunnel construction process, the deformation trend of the surrounding rock 01 along the construction direction of the tunnel is gradually changed and basically does not mutate; therefore, the installation position of the adjuster corresponding to the next construction section can be predicted by the third sub-model A-PINN according to the monitoring data of the current construction section.

[0060] ​It's important to understand that the arch frame 02 has multiple installation positions 021 along the tunnel cross-section. When installing the regulator, it's not necessary to install it at every installation position 021. For the deformation of the surrounding rock 01, placing the regulator precisely at the location of maximum deformation in the surrounding rock 01 yields the best deformation compensation effect. Therefore, the peak internal force point of the surrounding rock 01 can be obtained in the internal force field output by the second sub-model F-PINN. Since the installation position of the arch frame 02 and the location of the installation positions 021 are fixed, the installation positions 021 on the arch frame 02 may not correspond to the peak internal force point of the surrounding rock 01. Therefore, the regulator can only be installed at the installation position 021 closest to the peak internal force point to ensure that deformation compensation of the surrounding rock 01 is achieved with the minimum number of regulators.

[0061] Another aspect of this application provides a tunnel support structure internal force control system for implementing the aforementioned tunnel support structure internal force control method, such as... Figure 4 As shown, one preferred embodiment includes a displacement monitoring network, an active control network, and an edge analysis module 06. The displacement monitoring network is used to collect the cross-sectional deformation of the current construction section of the tunnel; the active control network compensates for the continuous internal force field of the current construction section based on the received internal force control signals; the edge analysis module 06 is connected to both the displacement monitoring network and the active control network, and is used to construct multiple sub-models based on the same neural network architecture. The edge analysis module 06 can receive the data collected by the displacement monitoring network, and then generate internal force control signals based on the constructed sub-models and send them to the active control network.

[0062] Specifically, such as Figure 4 As shown, the displacement monitoring network includes a displacement acquisition component 04 installed on the drainage board trolley 03 and a target 011 installed on the tunnel construction section; the displacement acquisition component 04 can move the drainage board trolley 03 to identify the target 011 installed on the current construction section.

[0063] Understandably, based on the number of monitoring points, the number of targets 011 can be set to five, spaced out along the midpoint of the cross-section and both sides. The displacement acquisition component 04 uses a camera or radar; the specific number of components 04 is determined according to the tunnel construction conditions. After the tunnel construction is completed and the initial support is erected, targets 011 are installed at predetermined positions in the surrounding rock 01, and the displacement changes of targets 011 reflect the local deformation of the surrounding rock 01. After data acquisition for each construction segment is completed, targets 011 can be removed for installation in the next segment.

[0064] Specifically, such as Figure 4 and Figure 5As shown, the active regulation network includes a hydraulic station 05 installed on the drainage plate trolley 03, and a regulator installed on the arch 02 required for the initial support construction. The regulator includes a support 1, a jacking part 2, a locking part 3 and a hydraulic loading assembly 4. The support 1 and the jacking part 2 can cooperate to form a jacking assembly, which is detachably installed on the arch 02. The jacking assembly can be telescoped within a set adjustment range, and mechanically locked by the locking part 3 when the set telescoping amount is reached. The hydraulic loading assembly 4 is detachably installed on the driving end of the jacking assembly to drive the jacking assembly to extend out; the hydraulic loading assembly 4 can be separated and detached from the jacking assembly after the jacking assembly is mechanically locked; the hydraulic station 05 can receive the internal force regulation signal of the edge analysis module 06, and the hydraulic station 05 can be connected with the hydraulic loading assembly 4 of the regulator through an oil circuit, so as to control the loading amount of the hydraulic loading assembly 4 according to the internal force regulation signal. For the specific structure of the regulator, please refer to the technical solution of the Chinese invention patent application with the publication number CN120004171A and the name of “A composite jacking device based on hydraulic loading and mechanical locking and a using method”.

[0065] It can be understood that the support 1 includes a fixedly connected base 11 and an outer sleeve 12; the base 11 can be fixedly installed on the arch 02 by means of bolts or welding. The jacking part 2 is slidingly installed in the outer sleeve 12, and is mainly used for jacking and supporting the surrounding rock 01 to achieve deformation compensation. The hydraulic loading assembly 4 includes a hydraulic cylinder 41 and a connecting assembly 42; the hydraulic cylinder 41 is detachably installed on the base 11 of the support 1 through the connecting assembly 42 and can be drivingly connected with the jacking part 2, so that when jacking is needed, the jacking part 2 is driven by the hydraulic cylinder 41 to slide along the support 1 to abut against the surrounding rock 01 through the oil output by the hydraulic station 05, and then a set jacking force is applied to the surrounding rock 01 according to the regulation scheme. The locking part 3 can be installed on the base 11 of the support 1 after the jacking part 2 is jacked to lock the position of the jacking part 2. The hydraulic loading assembly 4 can be removed from the support 1 after the locking part 3 is locked.

[0066] Specifically, the edge analysis module 06 uses the sub-models deployed in advance to calculate the displacement change of each monitoring point of the tunnel section at a fixed frequency (the frequency is determined according to the site conditions), and converts it into a global continuous displacement field through the first sub-model D-PINN, and analyzes the change of the internal force through the second sub-model F-PINN. After the displacement collection is completed, the latest global continuous displacement field and internal force field are read from the database, the third sub-model A-PINN for executing adjustment response prediction is called to design the adjustment load, and the adjustment load required for each preset adjustment position is converted into an oil circuit oil pressure signal, and then the hydraulic station 05 and the hydraulic loading assembly 4 of the regulator are connected, and the jacking part 2 of the regulator can be subjected to a corresponding jacking force through the corresponding oil circuit oil pressure signal.

[0067] The foregoing describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above-described embodiments, and the above-described embodiments and descriptions in the specification are only the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for regulating internal forces of a tunnel support structure, characterized by, The method comprises the following steps: S100: According to the structural characteristics of the tunnel and the primary support, a plurality of sub-models based on the same neural network architecture are constructed, including a first sub-model for performing global inversion of displacement, a second sub-model for performing internal force analysis, and a third sub-model for performing adjustment response prediction; S200: The tunnel is discretized in space and time to obtain a plurality of continuous construction sections; during the construction of the primary support in each construction section, the adjustors required by the active adjustment network and the targets required by the displacement monitoring network are installed at predetermined positions; S300: During the construction of the primary support in the entire construction section, the cross-sectional deformation of the current construction section is continuously collected through the displacement monitoring network, and the collected deformation data is input into the first sub-model to obtain the continuous displacement field of the current construction section; S400: The second sub-model analyzes the load distribution field obtained by the continuous displacement field inversion to obtain a continuous internal force field, and then the third sub-model generates a control scheme according to the continuous internal force field and predicts the predetermined position required for installing the adjustor in the next construction section; S500: The active adjustment network drives the adjustor installed at different predetermined positions in the current construction section according to the obtained control scheme.

2. The method of internal force regulation of a tunnel support structure according to claim 1, wherein, In step S100, each sub-model adopts a PINN architecture, and the specific construction includes the following processes: According to the structural parameters of the primary support material and the structural parameters of the tunnel, a control equation is constructed; The boundary of the control equation is given according to the construction method of the tunnel; Based on the combination of the DNN basic network model and the control equation and the boundary, a PINN basic model is obtained; A plurality of loss functions are set to optimize the PINN basic model, and based on the difference in the types of loss functions required for optimization, the required first sub-model, second sub-model and third sub-model are obtained.

3. The method of internal force regulation of a tunnel support structure according to claim 2, wherein, The loss function includes physical equation residual loss, boundary condition residual loss, initial condition residual loss and real collected data residual loss; The first sub-model is obtained by optimizing the PINN basic model through the physical equation residual loss, the boundary condition residual loss and the real collected data residual loss; The second sub-model and the third sub-model are obtained by optimizing the PINN basic model through the physical equation residual loss and the boundary condition residual loss, respectively, according to the different weights of the physical equation residual loss and the boundary condition residual loss.

4. The method of internal force regulation of a tunnel support structure according to claim 3, wherein, In the total loss function L corresponding to the first sub-model tol , the weights of the physical equation residual loss L phy and the real acquisition data residual loss L true are adaptive weights, and the specific acquisition process is as follows: With homoscedastic uncertainty, a trainable noise parameter is automatically learned for each loss function, and the L tol The expression is: ； where σ phy and σ true respectively represent the noise parameters corresponding to the physical equation residual loss and the real acquisition data residual loss, L bc represents the boundary condition residual loss, and λ bc represents the weight of the boundary condition residual loss.

5. The method of internal force regulation of a tunnel support structure according to claim 1, characterized in that, When installing the target, a plurality of points are set as the installation predetermined positions of the target in the cross-sectional direction of the current construction section of the tunnel at equal intervals; wherein the midpoint of the cross section is one of the point positions, and the remaining point positions are symmetrically distributed to both sides of the midpoint of the cross section.

6. The method of internal force regulation of a tunnel support structure according to claim 5, wherein, After the first sub-model obtains the deformation data of the predetermined positions collected by the displacement monitoring network, the deformation interpolation of adjacent predetermined positions in the cross-sectional direction of the construction section is performed to obtain the required continuous displacement field; When the construction length of the tunnel is less than a set value of the total length of the tunnel, an initial PINN basic model is trained through the measured data in the initial stage of the tunnel construction; the trained initial PINN basic model is connected in parallel with a traditional discriminator to construct a generative adversarial network discriminator as a physical discriminator; The control equation for constructing each sub-model is embedded into a generative adversarial network discriminator, and then in the process of constructing the continuous displacement field by the first sub-model, the generative adversarial network discriminator is used for overfitting suppression.

7. The method of tunnel support structure internal force regulation according to claim 1, wherein, A plurality of installation stations for installing the adjuster are arranged on the arch frame for erecting the primary support of the construction section at equal intervals. According to the continuous internal force field output by the second sub-model, the third sub-model outputs the internal force peak point of the continuous internal force field; and the installation station closest to the internal force peak point is selected as the installation scheduled position of the adjuster of the next construction section on the arch frame.

8. A system for regulating internal forces of a tunnel support structure, for implementing the method for regulating internal forces of a tunnel support structure according to any one of claims 1 to 7, characterized in that, It comprises: a displacement monitoring network; the displacement monitoring network is used for collecting the cross-section deformation of the current construction section of the tunnel; an active regulation network; the active regulation network compensates the continuous internal force field of the current construction section according to the received internal force regulation signal; and an edge analysis module; the edge analysis module is signal connected with the displacement monitoring network and the active regulation network respectively, and the edge analysis module is used for constructing a plurality of sub-models based on the same neural network architecture; the edge analysis module is adapted to receive the collected data of the displacement monitoring network, and then generate the internal force regulation signal according to the constructed sub-models and send it to the active regulation network.

9. The system for internal force regulation in tunnel support structures according to claim 8, characterized in that, The displacement monitoring network comprises a displacement collection component installed on a drain plate trolley and a target installed on the tunnel construction section; the displacement collection component is adapted to move the drain plate trolley, and then identify the target installed on the current construction section.

10. The system for internal force regulation in a tunnel support structure according to claim 8, characterized in that, The active regulation network comprises a hydraulic station installed on the drain plate trolley and an adjuster installed on the arch frame required for erecting the primary support; The adjuster comprises: a jacking assembly; the jacking assembly is detachably installed on the arch frame, and is adapted to stretch and contract within a set adjustment range and mechanically locked when reaching a set contraction amount; a hydraulic loading assembly; the hydraulic loading assembly is detachably installed on the driving end of the jacking assembly to drive the jacking assembly to stretch out; the hydraulic loading assembly is adapted to be separated and detached from the jacking assembly after the jacking assembly is mechanically locked; The hydraulic station is adapted to receive the internal force regulation signal of the edge analysis module, and is adapted to be oil connected with the hydraulic loading assembly of the adjuster, and then control the loading amount of the hydraulic loading assembly according to the internal force regulation signal.

Images