Engineering rock mass non-constitutive data driving calculation method and system

Abstract

The invention discloses an engineering rock mass non-constitutive data driving calculation method. The method comprises the following steps: acquiring geometrical characteristic parameters and spatial distribution rules of structural surfaces in various rock masses based on field geological survey; constructing a discrete fracture network model by adopting a layered and partitioned modeling method; respectively establishing databases of surface fractured rock mass and deep fractured rock mass; constructing a statically indeterminate equation for each data point in the database, and taking a least square method solution as a balance factor matrix; and simulating a stress field, a strain field and a displacement field of the engineering rock mass under the loading and unloading effects based on a non-constitutive data driving calculation method. The method is suitable for simulation analysis of the stress field and the deformation field of the engineering rock mass under external overload and excavation unloading. Aiming at the heterogeneous, discontinuous and anisotropic medium of the engineering rock mass, the method does not need to pre-define the constitutive relationship between the rock and the joint, can more accurately simulate the deformation behavior of the engineering rock mass under the complex geological condition, and provides an efficient and reliable technical means for slope excavation design and stability analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of data-driven computing, and particularly relates to a data-driven calculation method and system for engineering rock masses without constitutive models. Background Art

[0002] Due to long-term complex geological processes, discontinuities such as bedding planes, joints, faults, and fractures are commonly developed in rock masses. These structural planes are randomly distributed and intersect with each other in different scales, shapes, and directions, forming a complex network topology, which makes the rock mass exhibit significant heterogeneity, discontinuity, and anisotropy. Among them, the influence of structural planes on the mechanical properties of rock masses is much greater than that of intact rocks, mainly manifested as deformation under normal load and shear slip under tangential load. Therefore, accurately calculating the deformation and stress distribution of fractured rock masses under external loads is still a key scientific problem to be solved urgently.

[0003] In the analysis of fractured rock mass deformation, appropriate rock and joint constitutive models are needed to characterize the behaviors such as sliding, closing, and opening of structural planes under complex loads. However, due to the heterogeneity, discontinuity, and anisotropy of rock masses, existing models usually can only simulate the main mechanical properties of a certain type of rock mass under specific conditions, and it is difficult to comprehensively characterize the complex behaviors under arbitrary loads. In addition, the construction and verification of new models require a large amount of investment, and empirical assumptions may lead to error accumulation, limited application scope, and affect the reliability of test data.

[0004] In addition, in the actual engineering rock mass deformation analysis, when the number of cracks is small, joint elements can be used to explicitly describe the discontinuity of the rock mass, and it can be regarded as a collection of intact rock blocks separated by fractures. However, for areas with dense cracks, modeling each crack is neither practical nor will lead to a large model scale and low computational efficiency. Engineering experience shows that the surface cracks of rock slopes are significantly developed, the unloading opening is obvious, and it has a greater impact on the mechanical properties; while the internal rock mass is relatively intact, the cracks are closely combined, and the mechanical properties are better. Therefore, how to efficiently and accurately simulate the mechanical behavior of fractured rock masses has become a major challenge in engineering practice. Summary of the Invention

[0005] The present invention is made to solve the above problems, and aims to provide a data-driven calculation method for engineering rock masses without constitutive models, aiming to bypass the construction process of rock and joint constitutive models, and directly use rock stress-strain data points, joint normal stress-normal displacement data points, and tangential stress-tangential displacement data points, and seek the solution of the problem by minimizing the extreme value of the distance between these data points and the points that satisfy the conservation equations (equilibrium equations, geometric equations). This method can accurately capture the discontinuous deformation characteristics of the fracture network, so as to more realistically reflect the deformation characteristics of the rock mass.

[0006] The present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a calculation method for engineering rock masses without constitutive data driving, including the following steps:

[0008] Step 1: Based on on-site geological surveys, obtain the geometric characteristic parameters of structural planes in various rock masses and their spatial distribution laws.

[0009] Step 2: Stratify the rock masses based on the geometric characteristic parameters of structural planes in various rock masses and their spatial distribution laws, including surface fissured rock masses and deep fissured rock masses. Respectively model the surface fissured rock masses and deep fissured rock masses to obtain a global discrete fracture network model.

[0010] Step 3: Establish a rock database and a joint database for the surface fissured rock masses, and establish a stress-strain database of fissured rock masses under the REV size for the deep fissured rock masses.

[0011] Step 4: Construct a hyperstatic equation for each data point in the rock database, joint database, and stress-strain database of fissured rock masses, and use the least squares solution as the equilibrium factor matrix.

[0012] Step 5: Construct an optimization problem, solve the optimization problem based on the equilibrium factor matrix, and simulate the stress and deformation of engineering rock masses under loading and unloading.

[0013] Further, in step 2, a refined discrete fracture network model is used for the surface fissured rock masses; the mesoscopic fractures of the deep fissured rock masses are macroscopically characterized through the equivalent homogenization theory.

[0014] Further, in the global discrete fracture network model, the surface fissured rock masses are equivalent to a binary structure composed of solid elements and joint elements, and the deep fissured rock masses are equivalent to solid elements.

[0015] Further, in step 3, for the solid elements and joint elements of the surface fissured rock masses, an entity element database and a joint element database are respectively constructed; for the solid elements of the deep fissured rock masses, a stress-strain database of fissured rock masses under the REV size is generated based on the concept of the representative elementary volume REV.

[0016] Further, generating a stress-strain database of fissured rock masses under the REV size includes:

[0017] (1) Determine the representative element size of the deep fissured rock masses through numerical tests.

[0018] (2) For the deep fissured rock masses under the REV size, respectively use the entity element database and the joint element database. Through the homogenized strain boundary conditions, output the strain applied each time during loading and the calculated equivalent stress as a data point, thereby constructing a stress-strain database of fissured rock masses under the REV size.

[0019] Furthermore, the strain applied each time during loading and the calculated equivalent stress are output as a data point, including:

[0020] Assuming the rigid body displacement is 0, a preset macroscopic strain tensor is set, and node displacement constraints are applied to all nodes on the boundary according to the preset displacement function:

[0021] The equivalent stress tensor output is calculated by weighted calculation according to the integration weights of the material mass points:

[0022] Based on the preset macroscopic strain tensor and the output equivalent stress tensor as a data point, a stress-strain database of fractured rock mass under the REV size is constructed.

[0023] Furthermore, the hyperstatic equations constructed for each data point in the rock database, joint database, and stress-strain database of fractured rock mass in step 4 are:

[0024]

[0025] In the formula, min is the minimum value, N e and N je are the indexes of the solid element and joint element databases respectively; e data and je data are the total numbers of data points in the solid element and joint element databases respectively; and represent the stress-strain data pairs of the solid elements obtained from numerical tests; and represent the joint stress-displacement data pairs obtained from numerical tests; (C e ) ijkl and (C je ) ijkl are the equilibrium factor matrices of the solid element and joint element respectively.

[0026] Furthermore, in step 5, the coupled solution process of the material constitutive relation, equilibrium differential equation, and geometric compatibility equation in traditional mechanical analysis is converted into an optimization problem.

[0027] Furthermore, the optimization problem in step 5 includes a global optimization objective penalty function G and constraint conditions;

[0028] The global optimization objective penalty function G is:

[0029]

[0030] The constraint conditions include:

[0031]

[0032] In the formula, e = 1, 2, 3... n is the serial number of the numerical integration point of the solid element; je = 1, 2, 3... s is the serial number of the numerical integration point of the joint element; ω e is the element area integration weight of the numerical integration point of the solid element; ω je is the joint element length integration weight of the numerical integration point of the joint element; F e and F je are the spatial distances of the solid element and the joint element respectively; and are the strain and stress feasible solution component arrays at the numerical integration point of the solid element respectively; and are the stress and strain component arrays of the data distribution point of the solid element; and are the strain and stress feasible solution component arrays at the numerical integration point of the joint element respectively; and are the stress and strain component arrays of the data distribution point of the joint element; B e is the strain conversion matrix of the numerical integration point of the solid element; B je is the conversion matrix of the numerical integration point of the joint element in the global coordinate system; B' je is the conversion matrix of the numerical integration point of the joint element in the local coordinate system; T is the coordinate transformation matrix; u kl is the node displacement feasible solution vector array that satisfies the displacement boundary; f is the node external force load vector array; is the strain-stress database of the solid element; is the displacement-stress database of the joint element.

[0033] Second, the present invention provides an engineering rock mass constitutive data-free driven calculation system, including:

[0034] A parameter acquisition module, which is used to obtain the geometric characteristic parameters of the structural planes in various rock masses and their spatial distribution laws based on on-site geological surveys;

[0035] A network model construction module, which is used to layer the rock mass based on the geometric characteristic parameters of the structural planes in various rock masses and their spatial distribution laws, including surface fissured rock masses and deep fissured rock masses, and respectively model the surface fissured rock masses and deep fissured rock masses to obtain a global discrete fracture network model;

[0036] A database construction module, which is used to establish a rock database and a joint database for the surface fissured rock mass, and establish a stress-strain database of the fissured rock mass under the REV size for the deep fissured rock mass;

[0037] The balance factor matrix solving module is used to construct a hyperstatic equation for each data point in the rock database, the joint database, and the stress-strain database of fractured rock masses, and use the least squares solution as the balance factor matrix;

[0038] The optimization problem solving module is used to construct an optimization problem, solve the optimization problem based on the balance factor matrix, and simulate the stress and deformation of engineering rock masses under loading and unloading.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] (1) For the complex inhomogeneous, discontinuous, and anisotropic medium of engineering rock masses, the present invention can realize the stress-strain analysis under external loading and excavation unloading without defining its constitutive relationship, which has important application value.

[0041] (2) The present invention can accurately capture the discontinuous deformation characteristics of the fracture network, thus more realistically reflecting the deformation characteristics of rock masses.

[0042] (3) The present invention adopts a hierarchical and zonal modeling strategy to analyze engineering-scale problems, which significantly improves the calculation efficiency while ensuring the calculation accuracy, and provides an innovative solution for the numerical simulation of engineering-scale fractured rock masses. Description of the Drawings

[0043] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0044] Figure 1 It is a flowchart of a calculation method for engineering rock masses without constitutive data driven by the present invention.

[0045] Figure 2 It is a schematic diagram of the engineering rock mass model and dimensions of the embodiment of the present invention.

[0046] Figure 3 It is a schematic diagram of the grid calculation of fractured rock masses in the embodiment of the present invention.

[0047] Figure 4 It is the distribution of the rock database of the surface fractured rock masses (class V and class IV rock masses) in the strain space and stress space in the embodiment of the present invention.

[0048] Figure 5 It is the distribution of the joint database of the surface fractured rock masses (class V and class IV rock masses) in the displacement space and stress space in the embodiment of the present invention.

[0049] Figure 6 Schematic diagrams of calculation conditions with different sizes and orientations for the embodiments of the present invention.

[0050] Figure 7 Graph showing the relationship between the rock mass deformation modulus and the rock mass size with different orientations for the embodiments of the present invention.

[0051] Figure 8 Distribution of the rock database of deep fissured rock masses (class Ⅲ rock masses) for the embodiments of the present invention in the strain space and the stress space.

[0052] Figure 9 Displacement nephogram under self-weight calculated by the RS2 global model and the DDCM method for the embodiments of the present invention.

[0053] Figure 10 Displacement nephogram under excavation calculated by the RS2 global model and the DDCM method for the embodiments of the present invention. Detailed implementation manners

[0054] To make the above objects, features, and advantages of the present application more apparent and understandable, the following will describe the detailed implementation manners of the present application in conjunction with the accompanying drawings. Many specific details are set forth in the following description to facilitate a thorough understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the connotation of the present application. Therefore, the present application is not limited by the specific embodiments disclosed below.

[0055] Embodiment 1:

[0056] The two-dimensional numerical analysis model and the excavated body of a certain hydropower station slope in the embodiments of the present invention are as Figure 2 shown. The left and right boundaries of the model constrain the x-direction displacement, and the bottom boundary constrains the x-direction and y-direction displacements. In this embodiment, the displacement field information of the slope under self-weight and excavation is calculated by using the engineering rock mass constitutive data-driven calculation method provided by the present invention. The specific steps are as follows:

[0057] Step 1, according to the weathering degree, the slope rock mass is divided into class Ⅴ, class Ⅳ, and class Ⅲ rock masses from the outside to the inside (the parameters are shown in Table 1). Based on the geometric characteristics and spatial distribution characteristics of the structural planes in each type of rock mass obtained from geological surveys (the parameters are shown in Table 2), a fracture network is generated using the commercial software RS2.

[0058] Table 1

[0059]

[0060] Table 2

[0061]

[0062] Step 2, for the surface fractured rock masses of Class V and Class IV rock masses with large variations in mechanical properties, they are regarded as a binary structure composed of solid and joints, and the rock database and joint database are used for calculation; while for the fractured rock masses of Class III rock masses with relatively stable mechanical properties, based on the concept of equivalent continuous medium mechanics, the fractured rock masses are regarded as a composite material composed of rock and joints, and the database for generating equivalent fractured rock masses through the REV concept is used for calculation. The finally generated DDCM numerical model consists of 10282 solid elements and 1767 joint elements, as Figure 3 shown.

[0063] Step 3, databases for surface and deep fractured rock masses are established respectively. For the surface fractured rock masses (Class V and Class IV rock masses), based on the constitutive matrices of rock and joints and combined with appropriate strain ranges, a rock database and a joint database are generated, providing basic data support for subsequent data-driven calculations. Specifically, the rock database contains 20 3 samples, and each of the joint normal database and the tangential database contains 2000 samples. The strain ranges in the rock database are: ε x ∈[-5E-4,1E-3]; ε y ∈[-2E-3,0]; γ xy ∈[-1E-3,1E-3]. The normal displacement range in the joint normal database is: d n ∈[-3E-4,3E-4], and the tangential displacement range in the joint tangential database is: d s ∈[-1.5E-3,1.5E-3]. Figure 4 shows the distribution of the rock database in the strain space and stress space, where (a) is the strain data and (b) is the stress data; Figure 5 shows the distribution of the joint database in the displacement space and stress space, where (a) is the joint normal data and (b) is the joint tangential data.

[0064] For the deep fractured rock masses (Class III rock masses), an equivalent fractured rock mass database is constructed based on the representative elementary volume (REV) theory. First, the representative element size of the fractured rock mass is determined through numerical tests. The study investigated the equivalent deformation modulus of rock masses with side lengths of 10m, 20m, 30m, 40m, 50m, 60m, 70m, 80m, and 90m under axial loading in the directions of 0°, 30°, 60°, 90°, 120°, and 150°, as Figure 6 shown. It can be seen from the rose diagram (as Figure 7 shown) that when the size of the study area increases to 60m×60m and above, the axial equivalent deformation modulus in each loading direction tends to be stable, indicating that the representative element size has been reached.

[0065] Secondly, establish the stress-strain database of fractured rock mass under the REV size. For the fractured rock mass under the REV size, the rock database and the joint database are respectively used. By homogenizing the strain boundary conditions, the strain applied in each loading and the calculated equivalent stress are output as a data point, so as to establish the stress-strain database of the fractured rock mass under the REV size, and further used to drive the numerical simulation analysis of the deep fractured rock mass.

[0066] Assume that the rigid body displacement of the calculation model is 0, and apply the node displacement constraints to all nodes on the boundary according to the preset displacement function:

[0067]

[0068] In the formula, Γ represents the boundary region of the representative element; is the applied node displacement constraint; is the preset macroscopic strain tensor; x j is the coordinate vector. The output equivalent stress tensor is weighted according to the integration weight of the material material point.

[0069]

[0070] In the formula, A represents the area of the representative element; Ω represents the region of the representative element; ω represents the weight; σ represents the stress of each element.

[0071] By repeatedly applying different homogenized strain boundary conditions to the representative element, a total of 20 3 stress-strain data points are finally generated. The distribution of the strain of the data points in each component direction is: ε x ∈[-1E-3,1E-3], equally spaced loading (Δε x =1E-4); ε y ∈[-2E-3,5E-4], equally spaced loading (Δε y =1.25E-4); γ xy ∈[-2E-3,2E-3], equally spaced loading (Δγ xy =2E-4). The finally calculated equivalent stress-strain database of Class III rock mass is as Figure 8 shown, where (a) is the strain data and (b) is the stress data.

[0072] Step 4, in the data-driven calculation and solution, the selection method of the balance factor matrix C is: construct a statically indeterminate equation through the stress-strain data of each data point in the database, and use the least square solution as the balance factor matrix.

[0073]

[0074] where min is the minimum value, N e and N je are the indices of the entity element and joint element databases respectively; e data and je data are the total numbers of data points in the entity element and joint element databases respectively; and represent the stress-strain data pairs of entity elements obtained from numerical tests; and represent the joint stress-displacement data pairs obtained from numerical tests; (C e ) ijkl and (C je ) ijkl are the equilibrium factor matrices of entity elements and joint elements respectively.

[0075] Step 5: Based on the above-obtained Class V, IV, and III engineering rock mass databases, use the constitutive data-free driving algorithm to calculate the displacement nephogram of the engineering rock mass under self-weight and excavation. The global optimization objective penalty function G is:

[0076]

[0077] The constraint conditions include:

[0078]

[0079] In the formula, e = 1, 2, 3...n is the serial number of the numerical integration point of the entity element; je = 1, 2, 3...s is the serial number of the numerical integration point of the joint element; ω e is the unit area integration weight of the entity element integration point; ω je is the joint element length integration weight of the joint element integration point; F e and F je are the spatial distances of the entity element and the joint element respectively; and are the column arrays of the strain and stress feasible solution components at the entity element integration point respectively; and are the column arrays of the stress and strain components of the entity element assigned data points; and are the column arrays of the strain and stress feasible solution components at the joint element integration point respectively; and are the column arrays of the stress and strain components of the joint element assigned data points; B e is the strain transformation matrix of the entity element integration point; B je is the transformation matrix of the joint element integration point in the global coordinate system; B' jeis the transformation matrix of the integration point of the joint element in the local coordinate system; T is the coordinate transformation matrix; u kl is the column matrix of the feasible solution vectors of the nodal displacements that satisfy the displacement boundary; f is the column matrix of the nodal external force load vectors; is the strain-stress database of the solid element; is the displacement-stress database of the joint element.

[0080] Through the RS2 global model and the DDCM method, the displacement nephograms under self-weight action and excavation action are calculated respectively, as Figure 9 and Figure 10 shown. Figure 9 (a) and (b) are the displacement nephograms in the x and y directions under self-weight action calculated by the RS2 global model, Figure 9 (c) and (d) are the displacement nephograms in the x and y directions under self-weight action calculated by the DDCM method; Figure 10 (a) and (b) are the displacement nephograms in the x and y directions under excavation action calculated by the RS2 global model, Figure 10 (c) and (d) are the displacement nephograms in the x and y directions under excavation action calculated by the DDCM method. It can be seen from the figures that whether under self-weight action or excavation action, the displacement nephograms calculated by DDCM and RS2 show a high degree of consistency in the size and shape of the deformation area. In class V and class IV rock masses, the discontinuities of the rock mass (such as joints, fractures, etc.) have a significant impact on the deformation and stress distribution. The DDCM method can accurately simulate the influence of these discontinuities by using the rock database and the joint database to replace the traditional constitutive equation, making the calculated displacement nephogram highly consistent with the actual rock mass behavior. In class III rock masses, due to the relatively complete rock mass, the influence of discontinuities is relatively small. The DDCM method adopts the equivalent continuous medium method and uses the equivalent stress-strain database of the fractured rock mass to effectively capture the macroscopic mechanical behavior of the rock mass. By effectively simplifying the model structure and reducing the calculation complexity, this method significantly improves the calculation efficiency while ensuring the smoothness and reliability of the calculation results. Research shows that this method has important application value in the excavation design of engineering rock slope under complex geological conditions and can provide reliable theoretical basis and technical support for relevant engineering designs.

[0081] Embodiment 2:

[0082] This embodiment provides an engineering rock mass constitutive data-free driven calculation system, which is characterized by including:

[0083] A parameter acquisition module, which is used to obtain the geometric characteristic parameters of the structural planes in various rock masses and their spatial distribution laws based on on-site geological surveys;

[0084] The network model construction module is used to layer the rock mass based on the geometric characteristic parameters and their spatial distribution laws of the structural planes in various types of rock masses, including surface fissured rock masses and deep fissured rock masses, and respectively model the surface fissured rock masses and deep fissured rock masses to obtain a global discrete fracture network model;

[0085] The database construction module is used to establish a rock database and a joint database for the surface fissured rock mass, and establish a stress-strain database of the fissured rock mass under the REV size for the deep fissured rock mass;

[0086] The equilibrium factor matrix solving module is used to construct a hyperstatic equation for each data point in the rock database, the joint database and the stress-strain database of the fissured rock mass, and use the least squares solution as the equilibrium factor matrix;

[0087] The optimization problem solving module is used to construct an optimization problem, solve the optimization problem based on the equilibrium factor matrix, and simulate the stress and deformation of the engineering rock mass under loading and unloading.

[0088] As mentioned above, it is only the preferred specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in this application should be covered by the protection scope of this application.

[0089] It should be understood that the parts not detailed in this specification belong to the prior art.

[0090] It should be understood that the above description of the preferred embodiment is relatively detailed, and it should not be considered as a limitation to the protection scope of this invention patent. Under the inspiration of this invention, those of ordinary skill in the art can also make substitutions or deformations without departing from the scope protected by the claims of this invention, and all fall within the protection scope of this invention. The scope of protection requested by this invention shall be subject to the appended claims.

Claims

1. A calculation method for engineering rock masses without constitutive data-driven, characterized in that, It includes the following steps: Step 1: Obtain the geometric characteristic parameters of the structural planes in various rock masses and their spatial distribution laws based on on-site geological surveys; Step 2: Stratify the rock masses based on the geometric characteristic parameters of the structural planes in various rock masses and their spatial distribution laws, including surface fissured rock masses and deep fissured rock masses. Separate global discrete fracture network models are obtained by modeling the surface fissured rock masses and deep fissured rock masses respectively; Step 3: Establish a rock database and a joint database for the surface fissured rock masses, and establish a stress-strain database of fissured rock masses at the REV size for the deep fissured rock masses; Step 4: Construct a hyperstatic equation for each data point in the rock database, joint database, and stress-strain database of fissured rock masses, and use the least squares solution as the equilibrium factor matrix; Step 5: Construct an optimization problem, solve the optimization problem based on the equilibrium factor matrix, and simulate the stress and deformation of engineering rock masses under loading and unloading; 2. A calculation method for engineering rock masses without constitutive data-driven, according to claim 1, characterized in that In Step 2, refined discrete fracture network modeling is used for the surface fissured rock masses; the mesoscopic fractures of the deep fissured rock masses are macroscopically characterized through the equivalent homogenization theory; 3. The calculation method for engineering rock masses without constitutive data driving according to claim 2, wherein In the global discrete fracture network model, the surface fissured rock masses are equivalent to a binary structure composed of solid elements and joint elements, and the deep fissured rock masses are equivalent to solid elements; 4. The engineering rock mass constitutive data-driven calculation method according to claim 3, wherein, In Step 3, for the solid elements and joint elements of the surface fissured rock masses, a solid element database and a joint element database are constructed respectively; for the solid elements of the deep fissured rock masses, a stress-strain database of fissured rock masses at the REV size is generated based on the concept of the representative elementary volume REV; 5. A calculation method for engineering rock mass without constitutive data-driven, according to claim 4, characterized in that Generating the stress-strain database of fissured rock masses at the REV size includes: (1) Determine the representative element size of the deep fissured rock masses through numerical experiments; (2) For the deep fissured rock masses at the REV size, use the solid element database and joint element database respectively. Through the homogenized strain boundary conditions, the strain applied in each loading and the calculated equivalent stress are output as a data point, thereby constructing the stress-strain database of fissured rock masses at the REV size; 6. The calculation method for engineering rock mass without constitutive data driving according to claim 5, characterized in that The strain applied in each loading and the calculated equivalent stress being output as a data point includes: Assume that the rigid body displacement is 0, preset the macroscopic strain tensor, and apply node displacement constraints to all nodes on the boundary according to the preset displacement function; Output the equivalent stress tensor through weighted calculation according to the integration weight of the material mass points; Based on the preset macroscopic strain tensor and the output equivalent stress tensor being output as a data point, construct the stress-strain database of fissured rock masses at the REV size; 7. A calculation method for engineering rock masses without constitutive data driving according to claim 1, characterized in that: The hyperstatic equation constructed for each data point in the rock database, joint database, and stress-strain database of fissured rock masses in Step 4 is: where min is the minimum value, N e and N je are the indices of the solid element and joint element databases respectively; e data and je data are the total numbers of data points in the solid element and joint element databases respectively; and represent the stress-strain data pairs of the solid elements obtained from numerical tests; and represent the joint stress-displacement data pairs obtained from numerical tests; (C e ) ijkl and (C je ) ijkl are the balance factor matrices of the solid elements and joint elements respectively.

8. A calculation method for engineering rock masses without constitutive data-driven, according to claim 7, characterized in that: In Step 5, the coupling solution process of the material constitutive relationship, equilibrium differential equation, and geometric compatibility equation in traditional mechanical analysis is converted into an optimization problem; 9. The calculation method for engineering rock mass without constitutive data drive according to claim 8, characterized in that: The optimization problem in Step 5 includes the global optimization objective penalty function G and constraint conditions; The global optimization objective penalty function G is: The constraint conditions include: where \(e = 1, 2, 3,\cdots,n\) is the number of numerical integration points of the solid element; \(je = 1, 2, 3,\cdots,s\) is the number of numerical integration points of the joint element; \(\omega\) e is the element area integration weight of the solid element integration point; \(\omega\) je is the joint element length integration weight of the joint element integration point; \(F\) e and \(F\) je are the spatial distances of the solid element and the joint element, respectively; and are the column arrays of the feasible solution components of the strain and stress at the solid element integration point, respectively; and are the column arrays of the stress and strain components of the solid element assigned data points; and are the column arrays of the feasible solution components of the strain and stress at the joint element integration point, respectively; and are the column arrays of the stress and strain components of the joint element assigned data points; \(B\) e is the strain transformation matrix of the solid element integration point; \(B\) je is the transformation matrix of the joint element integration point in the global coordinate system; \(B'\) je is the transformation matrix of the joint element integration point in the local coordinate system; \(T\) is the coordinate transformation matrix; \(u\) kl is the column array of the node displacement feasible solution vector that satisfies the displacement boundary; \(f\) is the column array of the node external force load vector; is the strain-stress database of the solid element; is the displacement-stress database of the joint element.

10. An engineering rock mass constitutive data-driven calculation system, characterized in that, Include: A parameter acquisition module, which is used to obtain the geometric characteristic parameters and their spatial distribution laws of structural planes in various rock masses based on on-site geological surveys; A network model construction module, which is used to stratify rock masses based on the geometric characteristic parameters and their spatial distribution laws of structural planes in the various rock masses, including surface fissured rock masses and deep fissured rock masses, and respectively model the surface fissured rock masses and deep fissured rock masses to obtain a global discrete fracture network model; A database construction module, which is used to establish a rock database and a joint database for the surface fissured rock masses, and establish a stress-strain database of fissured rock masses under the REV size for the deep fissured rock masses; An equilibrium factor matrix solving module, which is used to construct a hyperstatic equation for each data point in the rock database, the joint database and the stress-strain database of fissured rock masses, and use the least squares solution as the equilibrium factor matrix; An optimization problem solving module, which is used to construct an optimization problem, solve the optimization problem based on the equilibrium factor matrix, and simulate the stress and deformation of engineering rock masses under loading and unloading; The engineering rock mass without constitutive data-driven calculation system executes the steps in the engineering rock mass without constitutive data-driven calculation method according to any one of claims 1-9.

Images