Tunnel geologic modeling method

Abstract

The invention discloses a tunnel geologic modeling method, and relates to the technical field of tunnel engineering and geologic modeling, and the method comprises the following steps: S1, collecting drilling data and lithologic spatial distribution information, calculating lithologic change frequencies and rock mass mechanical parameter fluctuation ranges of all rock stratums in the horizontal direction and the vertical direction, constructing a lithologic spatial variation scale index matrix, and calculating the lithologic spatial variation scale index matrix; the method is used for representing the spatial heterogeneity of the rock stratum. According to the method, geologically-driven space division is realized by introducing lithologic spatial variation scale indexes, and propagation of parameters in a structure continuous region is constrained by combining credible level control and boundary propagation limitation; and meanwhile, through multi-scale inversion verification and backtracking correction dynamic identification and modeling error optimization, a modeling process covering lithology identification, unit division, parameter control and closed-loop feedback is constructed, the precision and engineering applicability of a three-dimensional geologic model are effectively improved, and scientific support is provided for tunnel design and risk control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel engineering and geological modeling, and particularly relates to a tunnel geological modeling method. BACKGROUND

[0002] Tunnel geological modeling refers to constructing a digital model that truly reflects the underground geological structure along the tunnel line by using computer modeling technology based on multi-source data such as geological survey data, geophysical measurement results, drilling sampling analysis and construction feedback information in the process of tunnel design and construction. The model not only includes spatial structure information such as stratum distribution, lithological characteristics, fault fracture zone and underground water distribution, but also integrates geomechanical parameters and geological risk levels to realize the visualization expression of the geological environment and the prediction of the engineering behavior of the tunnel passing area. Through tunnel geological modeling, engineering personnel can optimize the tunneling scheme, identify risk sources and design the supporting structure in the design stage, and realize geological advanced prediction and dynamic adjustment in the construction stage, thereby improving the safety and economy of the tunnel engineering. It is the core foundation of the modern tunnel engineering informatization and intelligentization construction.

[0003] The prior art has the following disadvantages: In the existing three-dimensional tunnel geological modeling process, the vertical strip method is often used to divide the space of the modeling area, and the Kriging algorithm is used to perform spatial interpolation and assignment of rock mass mechanical parameters. However, if the strip scale selected does not match the actual rock mass spatial variation scale, structural modeling errors will be easily caused. When the strip division boundary passes through the lithological mutation zone existing in the interior of the geological body, the “non-physical boundary cutting” phenomenon will occur in the model, which will artificially cut off the lithological area that should be continuously transitioned, forming a “false strong zone” or a “false weak zone” that does not exist in the actual geology. On this basis, the Kriging interpolation algorithm fails to identify such boundary mutations, and its spatial correlation assumption will forcibly perform parameter transition interpolation in the lithologically discontinuous area, thereby producing phenomena such as “virtual jump” or “abnormal parameter diffusion” in the model, resulting in mechanical abnormal areas in the parameter field that do not conform to the actual engineering conditions. Such errors are difficult to detect in the conventional verification process in the modeling stage, but once the construction link is entered, it is extremely likely to cause serious mismatch between the tunnel support design and the actual rock mass strength, induce early instability of the surrounding rock, imbalance of support stiffness, tunnel over-excavation or local unloading expansion, and even cause large-scale continuous collapse or overall instability of the structure, bringing uncontrollable major risks to the engineering safety and progress.

[0004] The above information disclosed in the background section is only used to strengthen the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] The purpose of the present application is to provide a tunnel geological modeling method, by introducing a lithology spatial variation scale index, realizing geology-driven spatial division, and combining credible level control and boundary propagation restriction to constrain parameter propagation within a structure continuous region; at the same time, through multi-scale inversion verification and backtracking correction, dynamic identification and optimization of modeling errors are realized, a modeling process covering lithology identification, unit division, parameter control and closed-loop feedback is constructed, the accuracy and engineering applicability of the three-dimensional geological model are effectively improved, and scientific support is provided for tunnel design and risk control, so as to solve the problems in the above background art.

[0006] In order to achieve the above purpose, the present application provides the following technical scheme: a tunnel geological modeling method, comprising the following steps: S1, collecting borehole data and lithology spatial distribution information, calculating the lithology variation frequency and rock mass mechanical parameter fluctuation range of each rock layer in the horizontal direction and the vertical direction, and constructing a lithology spatial variation scale index matrix for representing the spatial heterogeneity of the rock layer; S2, according to the lithology spatial variation scale index matrix, performing spatial unit adaptive division on the modeling region, so that the division scale is consistent with the lithology variation scale, the lithology inside the unit is consistent, and the boundary avoids the lithology mutation position, forming a spatial unit structure consistent with the lithology distribution; S3, based on the spatial unit structure, evaluating the structural integrity and borehole data coverage of the unit, determining the credible level of each spatial unit, and setting parameter transmission constraint conditions according to the credible level to limit the parameter propagation between spatial units with a credible level meeting the requirements; S4, based on the parameter transmission constraint conditions, performing boundary control on the parameter propagation between spatial units, setting a propagation threshold at the boundary to block the parameter transmission between lithology discontinuous regions, and ensuring that the parameter only propagates inside the spatial units with similar lithology and a credible level meeting the requirements; S5, on the basis of completing the boundary propagation control, constructing a spatial unit parameter field, carrying out multi-scale inversion verification, identifying non-physical parameter mutation regions in the parameter gradient change obvious or lithology transition region, and performing backtracking analysis and structure correction on the abnormal regions; S6, based on the backtracking analysis and structure correction results, cooperatively optimizing the lithology spatial variation scale index matrix, the spatial unit division strategy and the parameter transmission constraint conditions, forming a closed-loop modeling process, realizing the control of structural modeling errors, and improving the accuracy and engineering applicability of the tunnel three-dimensional geological model.

[0007] Preferably, step S1 comprises: collecting borehole data and establishing a three-dimensional spatial database, and extracting the top and bottom interface positions of each rock layer; using hierarchical linear fitting to obtain the extension trend of the rock layer, and combining a local interpolation method to complete the interface; The lithology change frequency and the fluctuation range of the rock mass mechanical parameter are calculated to form a two-dimensional variation index table; A lithology spatial variation scale index matrix is formed by normalization processing and grade division, and a spatial unit division basis is determined by combining the variation field visualization technology.

[0008] Preferably, the step S2 comprises: A spatial division demand atlas is established based on the lithology spatial variation scale index matrix, and is divided into a high variation sensitive area, a medium variation transition area and a low variation stable area; A spatial unit division scale matching algorithm is implemented, and different division scales are adopted according to the variation grade; A weighted sliding scale window strategy is applied to adjust the unit boundary position, so that it avoids the lithology mutation zone; After the division is completed, consistency checking and borehole data coverage verification are performed to ensure that the spatial structure is consistent with the lithology distribution.

[0009] Preferably, the step S3 comprises: The structural integrity of the spatial unit is evaluated to identify whether there is a fault crossing, lithology mutation or structural deformity; The borehole data coverage degree is calculated to analyze the number of boreholes, the uniformity of distribution and the completeness of parameters; The structural integrity score and the data coverage score are weighted and combined to divide into a high confidence level, a medium confidence level and a low confidence level; According to the confidence level, a parameter transmission constraint condition is set, and a parameter propagation path control mechanism is constructed.

[0010] Preferably, the step S4 comprises: The lithology attribute relationship and the confidence level difference of the spatial unit boundary are identified to establish a boundary relationship matrix; According to the boundary relationship matrix, a propagation threshold rule is set to distinguish an open boundary, a semi-open boundary and a closed boundary; A spatial propagation path tracking algorithm is used to monitor the interpolation path in real time to determine whether the propagation threshold condition is met; The paths that do not meet the propagation condition are implemented to block the propagation or correct the interpolation, so as to ensure that the propagation process is controlled by the boundary limitation.

[0011] Preferably, the step S5 comprises: A spatial unit parameter field is constructed based on the parameter interpolation result under the boundary propagation control; The modeling space is divided into multiple scale sections, the parameter gradient change is calculated, and an abnormal response area is identified; The abnormal response area is analyzed in a backward direction to locate the error source; According to the backward analysis result, a structure is corrected, and a parameter field is reconstructed to verify whether the abnormality is effectively eliminated.

[0012] Preferably, step S6 comprises: correcting the lithology spatial variation scale index matrix based on the backtracking analysis result; optimizing and adjusting the spatial unit division strategy according to the corrected index matrix; reconfiguring the parameter transmission constraint conditions between the spatial units according to the structure adjustment result; reconstructing the parameter field and performing verification and evaluation to form a closed-loop modeling process.

[0013] In the above technical solution, the technical effects and advantages provided by the present application are as follows: The present application introduces a lithology spatial variation scale index matrix at the initial modeling stage, realizes the transition of spatial division from "geometry driving" to "geology driving", introduces the credible level control and boundary transmission restriction in the parameter propagation process, ensures that the interpolation process is subjected to the dual constraints of structure continuity and data integrity, prevents the mispropagation of parameters in the lithology discontinuous area, further realizes the dynamic identification and local optimization of model errors through the multi-scale inversion verification and backtracking correction mechanism. Finally, the method establishes a whole-process modeling system covering lithology identification, structure division, credible control, propagation restriction, anomaly detection and closed-loop optimization, significantly improves the expression accuracy and engineering adaptability of the model to complex geological structures, provides a more reliable geological basis for tunnel construction design and risk control, and has significant practical value and engineering popularization prospect. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments or prior art, the drawings needed to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0015] Figure 1 The method flowchart of the tunnel geological modeling method of the present application. DETAILED DESCRIPTION

[0016] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these example implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art.

[0017] The present application provides a tunnel geological modeling method as shown in Figure 1 The method comprises the following steps: S1, collecting drilling data and lithology spatial distribution information, calculating the lithology change frequency and the fluctuation range of rock mass mechanical parameters of each rock layer in the horizontal direction and the vertical direction, and constructing a lithology spatial variation scale index matrix based on the statistical results to represent the spatial heterogeneous characteristics of different rock layers; In order to ensure that the space division is highly consistent with the spatial characteristics of the true change of the geological body, the spatial variation scale of the rock mass is accurately extracted, and a data support basis for subsequent space division, i.e. a lithology spatial variation scale index matrix, is constructed. The process includes the following steps: Based on the drilling data obtained within the scope of the tunnel field, the system organizes the original information of each drilling hole coordinate, drilling depth, rock layer distribution, lithology classification, and rock mass mechanical parameters, and establishes a three-dimensional spatial database of drilling data. In order to accurately obtain the spatial distribution characteristics of the lithology, the rock layers appearing in each drilling hole are standardized and layered to ensure the consistency of the lithology code among different drilling holes. On this basis, the spatial position distribution of each rock layer in each drilling hole is extracted, and the spatial extension trend is preliminarily fitted combined with the three-dimensional geographic coordinates. In order to improve the accuracy of data processing, the layered linear fitting and local interpolation method are used to finely position the rock layer boundary, ensuring that the upper and lower limits of the rock layer have a clear and distinguishable extension path in space.

[0018] In three-dimensional geological modeling of tunnels, in order to accurately express the extension form and upper and lower limits of the rock layer in space, the top and bottom boundaries of the same rock layer in different drilling holes need to be spatially fitted and completed. The "layered linear fitting" and "local interpolation method" are the key means to achieve this purpose, each playing a different role and working together to complete the fine modeling of the rock layer boundary.

[0019] Layered linear fitting refers to the linear trend fitting of the upper and lower limit elevation data of the same rock layer appearing in multiple drilling holes in the horizontal direction (such as along the tunnel axis direction). Its role is to preliminarily restore the spatial trend and dip angle characteristics of the rock layer boundary on a large scale, i.e. to obtain the main extension trend of the rock layer in the overall space. Since most rock layers are controlled by sedimentation, structure, etc., they have a certain regularity of gentle dip or inclination on a regional scale, so linear fitting can better capture this trend.

[0020] The local interpolation method further compensates for the data gaps or elevation deviations in the local area due to geological disturbance or sparse drilling on the basis of the above linear trend. Common methods include inverse distance weighted interpolation (IDW), spline interpolation (Spline), or Kriging interpolation, etc. Its role is to locally estimate the missing area or boundary fluctuation according to the known boundary elevation points in the adjacent drilling holes, enhancing the spatial continuity and smoothness of the model.

[0021] Combining the two, the specific operation is: first, according to the stratification of the rock stratum, the top and bottom interfaces of each rock stratum are linearly fitted between the drilling data points to form a series of boundary line segments with preliminary inclination; Then, through local interpolation method, the boundary is supplemented and transitioned between the fitted line segments and the area without drilling, so that the boundary surface not only has the overall trend characteristics, but also has the local changes that conform to the real relief of geology. The final obtained rock stratum boundary surface not only has a clear identifiable path in three-dimensional space, but also can truly reflect the macrostructure and micro-perturbation of lithology distribution, greatly improving the modeling accuracy and the accuracy of geological interpretation.

[0022] After completing the fitting of the spatial boundary of the rock stratum, the lithology change frequency in the horizontal and vertical directions of each rock stratum is calculated. Specifically, equidistant sampling sections in the extension direction of the rock stratum are selected, and the lithology change points are counted on each section to determine the number of lithology category changes per unit length or unit depth, thereby obtaining the horizontal and vertical change frequencies reflecting the heterogeneity of lithology. At the same time, the mechanical parameters (such as uniaxial compressive strength, elastic modulus, Poisson's ratio, etc.) of each rock stratum in each borehole are normalized, and the numerical fluctuation range of each rock stratum in different spatial positions is analyzed, and the maximum-minimum value interval, mean value, standard deviation and coefficient of variation of each parameter in the corresponding rock stratum are calculated to quantitatively describe the dispersion degree of mechanical properties in space.

[0023] The lithology change frequency index and rock mass mechanical parameter fluctuation index obtained above are integrated, a two-dimensional lithology spatial variation scale index table is constructed, the corresponding change frequency value, parameter fluctuation range and fluctuation intensity of each rock stratum in the horizontal and vertical directions are recorded, and the spatial sensitivity factor is introduced by combining the coefficient of variation to form a unified scale expression. Further, through scale index normalization processing and grade division, the spatial variation scale of each rock stratum is divided into several grade intervals (such as high stability area, moderate variation area, strong variation area), and numerical labels are assigned for subsequent analysis. The scale index not only comprehensively reflects the spatial continuity and mechanical stability of the rock stratum, but also provides a quantitative control basis for spatial unit division in the subsequent modeling process, enhancing the geological adaptability of the modeling results.

[0024] Based on the lithology spatial variation scale index data formed by collation, the spatial interpolation analysis of the tunnel area three-dimensional geological field is carried out, the spatial thermal map or variation field visualization technology is adopted, the lithology change intensity and mechanical parameter discreteness are graphically expressed in three-dimensional space, and the change trend and sensitive interval of the rock mass properties in different regions are directly revealed. Combined with the tunnel axis and the position of the key section of construction, the variation scale index matrix is locally weighted, and the boundary transition zone of the spatial variation significant area and the relatively stable area is extracted, so as to lay a foundation for the subsequent adaptive division of the spatial unit. Through the above method, not only the fine extraction of the spatial variation law of the rock mass is realized, but also the precise quantitative basis for the subsequent core steps such as boundary control, parameter propagation constraint, error identification optimization is established, and the whole modeling process has high geological authenticity and engineering applicability in the data source stage.

[0025] The variation field visualization technology is a technical means for directly expressing the attribute change intensity (i.e. variation degree) of the geological body in three-dimensional space through graphical means, which is commonly used to reveal the distribution trend and discontinuity characteristics of lithology, mechanical parameters and the like in space. In the tunnel geological modeling process involved in the present application, the variation field visualization technology is mainly based on the lithology change frequency and the fluctuation degree of the rock mass mechanical parameters extracted in the early stage, the spatial attribute variation index is constructed, the variation intensity of each position is mapped into color gradient, isosurface or three-dimensional thermal map and the like, the direct presentation of the lithology spatial heterogeneity is realized. Its role is to assist in identifying the "high variation area", "weak variation area" and lithology transition zone in the tunnel surrounding rock, to provide visual discrimination basis for spatial unit division, and to be used as an important reference for subsequent parameter propagation control and model error identification. Through the technology, the engineering personnel can quickly judge in the three-dimensional environment which regions have sharp change and whether there is a potential mutation zone, so as to accurately set the boundary control strategy and parameter propagation threshold, and significantly improve the scientificity and engineering applicability of the modeling process.

[0026] The core role of this step is to establish a scientific, quantitative and geologically realistic spatial division basis for subsequent three-dimensional geological modeling, fundamentally solving the structural error problem caused by subjective setting of strip scale and ignoring the actual variation rule of lithology in traditional modeling. By systematically collecting drilling data and lithology spatial distribution information, the actual distribution state of the rock layer at different spatial positions can be obtained; further calculating the lithology change frequency of the rock layer in the horizontal and vertical directions can reveal the strength of the spatial heterogeneity of the lithology, and judge whether a certain area is relatively stable or mutates sharply. At the same time, combined with the statistical range of rock mass mechanical parameters (such as compressive strength, elastic modulus, etc.), the physical and mechanical stability and change trend of each rock layer in different spatial sections can be quantitatively reflected. The lithology spatial variation scale index matrix constructed based on these statistical results is actually a tool for unified expression of geological "structural change" and "engineering characteristics". It not only objectively reflects the spatial variation scale of different rock layers, but also can be used as the basis input condition for subsequent modeling steps such as spatial unit division, adaptive control, parameter propagation constraint, etc. In short, the implementation of this step will transform the experiential and subjective geological cognition into quantitative indicators with engineering operability, significantly improving the scientificity and precision of model division, and laying a key data foundation and logical starting point for realizing a three-dimensional geological model with high credibility and high adaptability.

[0027] S2, according to the lithology spatial variation scale index matrix, adaptively dividing the modeling area into spatial units, making the division scale of the spatial units consistent with the lithology spatial variation scale of the corresponding area, ensuring the high consistency of the lithology properties within each spatial unit, and avoiding the coincidence of the spatial unit boundary and the lithology mutation position, to establish a spatial unit structure consistent with the actual lithology distribution; In order to solve the problem of mismatch between strip scale and lithology change in traditional tunnel three-dimensional geological modeling, and avoid structural modeling errors caused by boundary mis-cut, a spatial unit adaptive division strategy based on the lithology spatial variation scale index matrix is adopted to establish a spatial division structure highly consistent with the actual lithology distribution. This process includes the following steps: According to the lithology spatial variation scale index matrix constructed in the preceding step, the entire tunnel modeling area is gridded according to the spatial coordinates, and the corresponding horizontal and vertical variation scale values of each coordinate sub-unit are obtained. Combined with the lithology change frequency and the fluctuation intensity of the mechanical parameters, a spatial division requirement map is established, i.e. identifying the areas with significant lithology variation intensity and areas with lower variation degree in the three-dimensional space, and dividing the space into high variation sensitive area, medium variation transition area and low variation stable area according to the variation intensity level, to provide target sections for subsequent unit scale matching.

[0028] On the basis of the division demand map, the spatial unit division scale matching algorithm is implemented. For high variation sensitive areas, small scale and high resolution spatial division strategy is adopted to ensure that the unit division granularity is fine enough to completely contain the lithology mutation area and avoid the boundary falling on the mutation point; for low variation stable area, larger scale division is adopted to improve the modeling efficiency without losing accuracy. In the division process, the progressive matching strategy based on the weighted sliding scale window is adopted to make the division boundary as much as possible inside the lithology continuous section, and dynamically adjust the boundary position to avoid the mutation zone and geological interface. The scale of all division units is directly driven by the local value in the lithology variation index matrix without artificial intervention, which fundamentally avoids the error caused by subjective division.

[0029] The "spatial unit division scale matching algorithm" refers to adaptively determining the division scale (i.e. size and distribution density) of each spatial unit according to the lithology spatial variation characteristics of different geological areas, to realize the accurate correspondence between the modeling unit and the actual variation degree of the rock mass. In this step, the role of this algorithm is to break the limitations of the traditional fixed scale division method, so that the spatial division scale is no longer one-size-fits-all, but dynamically responds to the severity of lithology change: in areas where lithology mutates frequently and mechanical parameters fluctuate sharply, small unit division with fine granularity is adopted to accurately capture complex structural details; while in areas where lithology is continuous and stable, and mechanical properties are uniform, large scale division is adopted to improve modeling efficiency and simplify parameter processing, thus achieving an optimal balance between accuracy and efficiency.

[0030] The implementation process of the algorithm specifically includes the following key steps According to the lithology spatial variation scale index matrix, the variation intensity level of each division area in the horizontal and vertical directions is obtained; According to the preset scale mapping rule, the high variation level area is corresponded to a smaller division scale (such as 1 meter to 3 meters), and the low variation level area is corresponded to a larger division scale (such as 5 meters to 10 meters); In the division process, the "weighted sliding scale window" strategy is adopted, that is, the variation level is taken as the weight, the variation gradient of adjacent areas is analyzed by sliding, and the unit boundary is dynamically adjusted to avoid the variation jump point; Through progressive matching and local correction, the consistency of the unit internal lithology is realized while ensuring that the boundary does not cross the mutation zone. The algorithm is essentially a scale regulation mechanism based on geological attribute driving, which not only improves the scientificity and intelligence of the division, but also completely avoids the structural error caused by subjective and experienced division, providing a solid foundation for building a real and reliable three-dimensional geological model.

[0031] After the initial adaptive division, the consistency of lithological properties between adjacent spatial units is checked according to the principle of lithological structure consistency. If a high-frequency mutation of lithological properties is found between two adjacent units, the boundary position is adjusted by retracing to move the boundary to a position where the lithological change tends to be stable. This process automatically evaluates and locally optimizes the rationality of the boundary transition by iteratively comparing the mean and standard deviation of the lithological code in each spatial unit. This ensures that the lithology within each spatial unit is highly consistent, the boundary avoids abrupt boundaries as much as possible, and spatial continuity is achieved.

[0032] After completing the adaptive division of spatial units, the resulting spatial division results are mapped one by one with the original drilling data to ensure that all drilling information can be completely covered within a spatial unit. Through the coverage evaluation model, the division accuracy and data positioning are verified again to identify whether there are data missing or duplication problems caused by too small scale or mispositioned division boundaries. If there are abnormal coverage areas, fine-tune the division again to form a final set of spatial unit structures that can be used for modeling. This division structure not only fully reflects the actual distribution and variation characteristics of lithology in space, but also provides a structural basis for subsequent parameter transmission control and error checking, fundamentally achieving precise alignment and error elimination of the geological model in the spatial structure level.

[0033] The role of this step is to guide the division of spatial units in the tunnel three-dimensional geological model based on the quantitative results of lithological spatial variation scale, achieving high consistency between the model spatial structure and the real distribution of geology, and fundamentally avoiding modeling errors caused by subjective division scale setting or uniform strip cutting. In traditional modeling methods, a fixed scale strip method is often used, regardless of whether the lithology changes dramatically or not, the spatial block is cut according to a uniform size, which ignores the heterogeneous characteristics of rock layers in space, and easily leads to the boundary of spatial units crossing the lithological mutation zone, forming "non-physical cutting". The "adaptive division of spatial units based on variation scale" implemented by this step can dynamically adapt the scale of spatial units to the degree of lithological change: small-scale division is used in areas with strong lithological mutation and large parameter fluctuations to accurately capture complex structures; large-scale division is used in areas with stable lithology and slow changes to improve modeling efficiency and ensure parameter stability. This on-demand adjustment and responsive division method not only ensures the high consistency of lithology within the unit, but also maximizes the avoidance of boundary overlap with lithological mutation zones, eliminating the causes of "false strong bands" and "false weak bands" from the structure, effectively controlling structural errors in the modeling stage, and significantly improving the geological matching degree and the rationality of mechanical parameter distribution of the model. At the same time, this step provides a precise spatial partition basis for subsequent parameter propagation constraints, boundary blocking processing, and error identification and feedback correction, and is an important hub link in the full-process closed-loop modeling.

[0034] S3, based on the established spatial unit structure, evaluating the structural integrity and drilling data coverage of each spatial unit, determining the trust level of each spatial unit combined with the evaluation results, and setting parameter transmission constraint conditions between spatial units according to the trust level to limit the propagation of parameters only between spatial units whose trust level meets the requirements; In order to ensure that the propagation of rock mass mechanical parameters in the modeling space has geological rationality and data support basis, based on the constructed spatial unit structure, the structural integrity and drilling data coverage of each spatial unit are quantitatively evaluated, and the trust level of each spatial unit is further determined, and the parameter transmission constraint conditions between spatial units are set according to the trust level, so as to effectively control the path and range of parameter propagation. The process includes the following steps: The structural integrity of the divided spatial unit is evaluated. Specifically, it includes: calculating whether there is a fault crossing, a sudden change of lithology or a strong discontinuity feature within the boundary range of each spatial unit; statistics whether the lithology attribute inside the unit has continuity, consistency and closure; analyze whether there are problems such as unclear boundary, local inlay, structural deformity in the spatial unit. For spatial units with complete structure and homogeneous lithology, mark them as structural integrity units; for spatial units with unclear boundaries or complex lithology changes, mark them as structural anomaly units. This evaluation ensures that each spatial unit not only has geometric independence, but also has the ability of continuous expression of lithology and physical characteristics, providing a structural basis for subsequent trust level determination.

[0035] The coverage of drilling data for each spatial unit is calculated. The specific method includes: counting the number of drillings contained in the unit, the uniform distribution of drilling points in space, the types of lithology and the completeness of mechanical parameters involved in the drilling data. Further analyze the support strength of drilling data on the lithology and parameter characteristics of the spatial unit, such as whether it can represent all the main rock layers of the unit, whether there is a data blind area or a problem of insufficient representation. The coverage is divided into three categories: high coverage, partial coverage and invalid coverage, and each category is configured with a standardized score. Through this coverage evaluation, it is ensured that there is enough data support inside the spatial unit, so as to guarantee the effectiveness and credibility of subsequent interpolation or parameter transmission.

[0036] Based on the above two evaluation results, the credibility level of each spatial unit is determined. The credibility level is defined as the comprehensive reliability evaluation result of the spatial unit in terms of lithological structure expression and data support. The structural integrity score and the data coverage score are combined by weighting to divide into three level intervals of high credibility level, medium credibility level and low credibility level. High credibility level units represent clear structure, homogeneous lithology and sufficient data support, which are suitable for parameter propagation source and path; medium credibility level units have certain structure and data support, but there is a certain uncertainty; low credibility level units are usually complex structure, poor data coverage and dramatic change area, which are not suitable for propagation path or reference basis.

[0037] According to the credibility level of the spatial unit, the parameter transmission constraint condition is determined to control the propagation range and path of rock mass mechanical parameters. The specific way is: allow parameters to propagate freely between high credibility level spatial units, set propagation weight attenuation mechanism when propagating from high credibility level units to medium credibility level units to avoid information distortion; prohibit parameters to directly cross to low credibility level units or propagate between such units to prevent false propagation path from causing model distortion. At the same time, parameter blocking conditions are set at boundaries with large credibility level differences to limit the interpolation algorithm from including low credibility areas into high credibility results, thereby forming a propagation closed loop control mechanism. In this way, a parameter transmission control framework based on credibility level is effectively constructed, which ensures that the propagation of parameters in space follows the geological structure and data logic, prevents misleading diffusion and non-physical jumps, and provides support and protection for building accurate and reliable three-dimensional geomechanical models.

[0038] The role of this step is to introduce a parameter reliability evaluation mechanism based on data support and structural rationality in the process of three-dimensional geological modeling, to establish a reasonable and controllable path constraint for the subsequent spatial propagation of rock mass mechanical parameters, and to prevent misleading expansion of interpolation algorithms in areas of geological uncertainty. In traditional modeling methods, spatial interpolation techniques such as Kriging algorithm usually propagate parameter information without distinction in the entire modeling area, ignoring the fact that some areas may have missing borehole data, structural distortion, and lithological mutations. This uncontrolled propagation mode is prone to cause non-physical errors such as "parameter jump" and "false diffusion" in areas of low data density or complex geological structures. This step systematically evaluates each spatial unit from two dimensions: first, the integrity of its geometry and lithological structure, i.e., whether the unit has consistent lithology and clear boundaries; second, the degree of borehole data coverage, i.e., whether it has enough data to support its parameter expression. Based on the above dual evaluation results, each spatial unit is assigned a reliability level, and clear transmission conditions and boundary restrictions are set during parameter propagation. For example, high-reliability spatial units can continuously transmit parameters, while low-reliability areas restrict or block parameter propagation, cutting off potential error chains from the propagation path. This mechanism not only significantly improves the realism and stability of the model results in spatial distribution, but also provides a reliable foundation for subsequent error identification and optimization feedback of the model structure, and is a key control link to ensure the applicability and scientific effectiveness of the model engineering.

[0039] S4, based on the set parameter transmission constraint conditions, boundary control of parameter propagation between spatial units is performed, a propagation threshold limit is set at the boundary of the spatial unit, the parameter transmission channel between lithologically discontinuous areas is blocked, it is ensured that the parameter only propagates within spatial units with similar lithology and meet the requirements of reliability level, and abnormal parameter diffusion is prevented; To prevent non-physical diffusion or jump of rock mass mechanical parameters during spatial propagation across lithologically discontinuous areas or low-reliability areas, a boundary control strategy based on parameter transmission constraint conditions is proposed. This strategy actively blocks propagation paths that do not have geological continuity or data support conditions by setting a propagation threshold, ensuring that parameters only propagate within spatial units with similar lithology and meet the requirements of reliability level, thereby improving modeling accuracy and reducing interpolation errors. This process includes the following steps: Establish the spatial unit boundary attribute identification mechanism. On the basis of the completed credibility level division and lithology consistency determination, the boundary surface of each spatial unit is extracted, and the lithology attribute comparison relationship and credibility level difference relationship with the adjacent spatial unit are identified. The lithology attribute comparison includes lithology category consistency, structure evolution trend, parameter change direction and the like; the credibility level difference includes the same level, level decrease or level fault and the like. Through the above identification process, a "boundary relationship matrix" is constructed, and whether there is lithology discontinuity or credibility level difference on each boundary is determined, which provides boundary condition basis for subsequent propagation control.

[0040] According to the boundary relationship matrix, the propagation threshold rule is set. For the boundary with consistent lithology and equal credibility level, the boundary is set as "open boundary", and the parameters are allowed to continuously propagate between the units; for the boundary with consistent lithology but different credibility levels, the boundary is set as "semi-open boundary", and the parameters can propagate but are subjected to weight attenuation processing to reduce the transmission uncertainty; for the boundary with inconsistent lithology or the boundary with serious mismatch of credibility level, the boundary is set as "closed boundary", and any parameter cross-transmission behavior is blocked. In the actual modeling process, the propagation threshold is set as a control factor, and when the lithology difference index or the credibility level difference index exceeds the upper limit of the factor, the blocking mechanism is automatically triggered, thereby forming a selective propagation barrier.

[0041] The parameter diffusion behavior is dynamically monitored in the boundary propagation process. The spatial propagation path tracking algorithm is used to analyze the boundary type and corresponding threshold state of the propagation path in real time in each parameter interpolation process. Once it is identified that the propagation path is about to cross the closed boundary, the propagation calculation on the path is terminated immediately; if the path crosses the semi-open boundary, the interpolation result is modified according to the weight attenuation factor, so as to ensure that the final transmission result meets the boundary constraint logic. At the same time, the path, boundary type and propagation response state of each propagation behavior are recorded to form a propagation history record, which is used for subsequent modeling consistency checking and error identification tracking.

[0042] The spatial propagation path tracking algorithm refers to recording and analyzing the propagation path between each interpolation value point and its sampling source point in space in real time in the rock mass parameter interpolation process, and then judging whether the path crosses the boundary of different types of spatial units, and deciding whether to allow the parameter propagation on the path to continue or to be modified according to the attribute of the crossed boundary and the threshold limit. In the present application, the algorithm gives the parameter interpolation process the "spatial perception" ability, so that each interpolation behavior is no longer a pure mathematical operation based on distance weight, but a controlled propagation based on "whether it meets the geological structure and credibility level conditions".

[0043] The implementation process is: first, identify the sample point set associated with the target interpolation point when performing parameter interpolation each time; then, decompose the three-dimensional space path between each pair of "interpolation point-sampling point" into segments, and determine whether the path crosses one or more space unit boundaries; for each boundary segment, call the pre-established boundary relationship matrix to identify its lithology continuity, confidence level relationship and preset propagation threshold. If a path passes through an open boundary, the interpolation proceeds normally; if the path passes through a semi-open boundary, an attenuation factor is applied to adjust the interpolation contribution; if the path passes through a closed boundary, the interpolation contribution is set to zero, and the influence of the path on the interpolation point is automatically terminated. This algorithm runs in real time in a point-by-point tracking manner during the interpolation process, ensuring that each parameter transfer is not only constrained by spatial distance, but also by geological structure and information reliability, thereby preventing non-physical jump propagation of parameters in areas of discontinuous lithology or low confidence, effectively improving model interpolation accuracy and spatial rationality.

[0044] The parameter propagation process in the entire modeling area is comprehensively reviewed to ensure that all propagation paths meet the boundary control logic and do not exhibit unauthorized diffusion behavior. By comparing the propagation path records with the boundary relationship matrix, it is analyzed whether there are abnormal situations such as jump propagation, fault boundary crossing propagation or closed boundary failure propagation. If a boundary control failure area is found, the control logic is systematically traced back, the propagation rules are corrected or the boundary identification results are adjusted, forming a closed-loop optimization process for propagation control.

[0045] The purpose of this step is to establish a parameter propagation boundary control mechanism based on geological structure consistency and data reliability in the process of three-dimensional geological modeling, to avoid non-physical jump propagation or abnormal diffusion of rock mass mechanical parameters in space from the root cause, and to improve the spatial expression accuracy and engineering applicability of the model. In the traditional modeling process, interpolation algorithms (such as Kriging interpolation) usually assume that the entire modeling area is a spatially continuous and smooth attribute field, ignoring the existence of lithological discontinuous interfaces, tectonic fault zones and data sparse areas. This unconstrained interpolation can easily lead to non-discriminatory propagation of parameters on both sides of geological mutation zones, resulting in the generation of "false strong bands" or "false weak bands" in the model that do not conform to geological reality, and further affecting tunnel support design and construction risk prediction.

[0046] By this step, the propagation threshold restriction mechanism is systematically introduced on the boundary of spatial units, and it is explicitly stated that the parameter transfer should be actively blocked or weakened between spatial units with different lithology discontinuity, parameter fluctuation, or significant difference in confidence level. Specifically, by boundary type identification and threshold setting, a "open-semi-open-closed" three-class boundary system is constructed to control the path and strength of parameter propagation. In the interpolation execution phase, with the help of spatial propagation path tracking algorithm, it is judged in real time whether the interpolation path crosses the boundary that does not allow propagation, and accordingly the propagation interruption or weight adjustment is implemented. This process not only improves the spatial intelligence and selectivity of the interpolation process, but also effectively prevents the "wrong diffusion" of parameters from high-confidence areas to low-confidence areas, ensuring that the mechanical parameter distribution in the modeling result is closer to the actual geological conditions. Overall, this step is one of the key links for the invention to achieve high-precision modeling. It transforms the original mathematical interpolation behavior based on spatial distance into a controlled propagation process that takes into account geological constraints and data quality, and actively prevents and controls structural errors and intelligently selects propagation paths.

[0047] S5, on the basis of completing the boundary propagation control, the spatial unit parameter field is constructed, and multi-scale inversion verification is carried out, the parameter gradient change obvious area and the lithology transition area are selected as the key detection object, the possible non-physical parameter mutation area between spatial units is identified, and the abnormal area is analyzed and the structure is corrected; To further improve the modeling accuracy, verify the rationality of parameter propagation, and actively identify possible structural errors, after completing the boundary control based on the propagation threshold restriction, the spatial unit parameter field is constructed, and multi-scale inversion verification analysis is carried out, the potential non-physical parameter mutation area between spatial units is identified, and through backtracking analysis and structural correction, the model is iteratively optimized. The process includes the following steps: Based on the parameter interpolation results after completing the boundary propagation control, the three-dimensional parameter field is constructed according to the spatial unit. Specifically, the rock mass mechanical parameters (such as elastic modulus, compressive strength, Poisson's ratio, etc.) in each spatial unit are uniformly coded and spatially mapped after interpolation, generating a continuous parameter distribution body, forming a parameter field expression with spatial topological structure. This parameter field not only records the parameter value of each spatial unit, but also retains its boundary propagation path, confidence level, lithology matching information and other attributes as the input basis for subsequent multi-scale verification.

[0048] Multi-level parameter field verification based on scale partition is carried out. According to the structural complexity and lithology distribution characteristics of the tunnel area, the entire modeling space is divided into several multi-scale detection sections, including the whole macro-scale area (used to identify the overall trend of migration), the local meso-scale area (used to analyze the continuity of the partition), and the micro-local scale area (used to monitor the abnormal point). In each scale area, the gradient distribution of the spatial parameter is calculated, and by comparing the actual lithology transition zone position with the parameter change trend, those "abnormal response areas" where the parameter mutation is severe but the lithology does not change, or the lithology transitions but the parameter does not respond, are identified, and the possible modeling error positions are preliminarily locked.

[0049] The identified abnormal response areas are analyzed backtracking, and the causes are tracked one by one. The analysis path includes: checking whether the parameter propagation path of the area passes through the boundary of the space unit with large difference in credibility level; judging whether the propagation has violated the propagation threshold limit condition or there is abnormal weight superposition behavior; comparing the original drilling data to check the error amplitude between the interpolation result and the sampling value; checking whether there are potential problems such as lithology classification error, boundary division offset, or local unit structure deformity in the area. Through piece-by-piece verification and data comparison, the root cause of abnormal mutation is determined, and the basis for the next step of structure correction is provided.

[0050] According to the results of backtracking analysis, the structure of the related abnormal area is corrected. If it is found that the propagation path is incorrect due to improper division of the unit boundary, the position of the space unit boundary is adjusted to avoid the mutation point again; if it is found that the propagation right is misjudged due to distortion of the credibility level evaluation, the drilling data coverage and structure consistency of the area are re-evaluated, and the credibility level distribution is corrected; if the propagation weight is abnormal or the interpolation method is invalid, the interpolation algorithm is replaced locally or the propagation threshold is reset to limit the interference range of the parameter. After the correction is completed, the parameter field of the updated area is reconstructed and verified again until the abnormal response area is effectively eliminated and the overall structure of the model tends to be stable.

[0051] The purpose of this step is to introduce a systematic verification and feedback mechanism after modeling is completed. By constructing a spatial unit parameter field and combining multi-scale inversion analysis, non-physical parameter mutation phenomena that may occur during modeling can be effectively identified and corrected, ensuring that the model is highly consistent with the actual structure and has engineering interpretation and stability in numerical value. In the traditional three-dimensional geological modeling process, parameter interpolation and model structure division are usually considered as the end point, and there is a lack of systematic post-diagnosis means, which can easily lead to local "jumping", "spike" or "weak zone" and other phenomena that do not conform to the actual geology due to unreasonable boundary setting, misleading propagation path, and abnormal parameter superposition. Through this step, a complete rock mass mechanical parameter three-dimensional distribution body is first constructed based on spatial units, and on this basis, a multi-scale inversion analysis method is introduced to test the entire modeling area from macro to micro: the macro level is used to identify large-scale trend deviation; the meso level is used to detect parameter consistency in transition areas; and the micro level precisely locates local mutation points. This multi-scale collaborative analysis method can sensitively capture abnormal areas caused by propagation errors, boundary errors, or limitations of interpolation algorithms. After identifying abnormal areas, the error sources are analyzed one by one by tracing their propagation path, setting their confidence level, checking their original data coverage, and adjusting their boundary division logic. Spatial unit boundary adjustment, parameter propagation control correction, or local interpolation reconstruction are then performed to achieve secondary optimization of the modeling structure and parameter field. This step not only improves the reliability and error resistance of the model results, but also establishes a closed-loop modeling quality assurance mechanism, providing more reliable geological basis for support design, construction control, and risk warning in tunnel engineering. It is a key link for realizing the transition from "building" to "credibility" of engineering-level three-dimensional geological models.

[0052] S6, based on the results of backtracking analysis and structure correction, the lithology spatial variation scale index matrix, spatial unit division strategy and parameter transmission constraint condition are optimized collaboratively to form a closed-loop modeling process from lithology variation scale extraction, spatial unit division, confidence level control, parameter propagation restriction, abnormality detection and verification to feedback correction, effectively control the structural modeling error, and improve the accuracy and engineering applicability of the tunnel three-dimensional geological model; To realize the structural closed-loop control in the process of three-dimensional geological modeling of tunnels, ensure the consistency and dynamic collaboration between lithology structure division, parameter propagation logic and error response mechanism, a collaborative optimization method based on the results of backtracking analysis and structure correction is proposed to update the lithology spatial variation scale index matrix, spatial unit division strategy and parameter transmission constraint condition, and then build a complete modeling closed-loop process integrating extraction-division-control-propagation-verification-optimization. The process includes the following steps: According to the backtracking analysis results of the abnormal area, the lithology spatial variation scale index matrix is updated and corrected. The analysis found that the abnormal area is usually accompanied by problems such as insufficient statistical frequency of original lithology change, low estimation of parameter fluctuation intensity, or too coarse spatial scale division. Therefore, the high gradient mutation area identified in the inversion verification process is re-included in the lithology variation statistics, and its weight is increased. At the same time, the variation frequency and parameter fluctuation interval of the corresponding area are recalculated through an enhanced local interpolation method. The updated variation scale index matrix is normalized to correct the original lithology variation expression accuracy, making it more truly reflect the spatial distribution of geological heterogeneity and provide more refined scale basis for subsequent spatial division.

[0053] Based on the updated lithology spatial variation scale index matrix, the spatial unit division strategy is optimized simultaneously. For the areas that have exposed boundary misplacement, lithology mismatch within the unit, or structural deformity in the previous modeling process, the variation scale level is re-analyzed, and a small-scale subdivision strategy is used to improve its spatial resolution. For the areas that are over-divided and have stable lithology, the unit scale is appropriately merged or expanded to improve the modeling efficiency and parameter stability. In the division process, a dynamic window adjustment method based on spatial sensitivity factor is used to realize real-time fine tuning of the boundary position, ensuring that the boundary falls as much as possible within the lithology continuous area, avoiding the formation of "cutting mutation zone" structural error again, and improving the structural rationality of the whole modeling space.

[0054] Based on the structure-adjusted spatial unit, the reliability level of each unit is re-evaluated, and the parameter transmission constraint conditions are updated simultaneously according to the reliability level. For the areas that were underestimated in reliability level but supported and enhanced by the supplementary data and reconstructed data, their reliability level is appropriately improved to make them have the right to spread; while for the areas with complex structure and insufficient data update, their reliability level is maintained or lowered to avoid new error propagation. At the same time, the boundary threshold setting rules are updated at the propagation logic level, and the judgment conditions of "open boundary", "semi-open boundary" and "closed boundary" are re-set for the optimized unit relationship, to ensure that the propagation path is more consistent with the latest geological structure and data distribution, and to improve the rationality and safety of the parameter transmission process.

[0055] After the above-mentioned index, division and propagation logic are cooperatively optimized, the new round of parameter interpolation results are reconstructed into a spatial parameter field, and multi-scale verification analysis is carried out again. By differentiating the new parameter field from the original model, the structural stability of the optimized model, the error change trend and the newly generated abnormal area are identified, and the cooperative optimization effect is further evaluated. If the verification result shows that most of the original abnormalities have been suppressed, the parameter field continuity has been enhanced, and the model response stability has been improved, the optimization is confirmed to be effective; if there are still new local inconsistency areas, the next round of feedback correction is continued. Through such a cycle iteration, a closed-loop modeling process composed of lithology extraction, unit division, grade evaluation, propagation control, abnormal verification and result optimization is finally formed, which not only dynamically eliminates structural modeling errors, but also significantly improves the adaptability of the tunnel three-dimensional geological model in complex geological environment, the engineering guidance value and the long-term stability.

[0056] The role of this step is to build a dynamic and self-correcting closed-loop modeling mechanism. Through the cooperative optimization of key links in the geological modeling process, such as lithology variation scale extraction, spatial unit division, parameter propagation control and other elements, the structural modeling error problems caused by unreasonable local structure division, out-of-control parameter propagation path and mismatched data reliability are solved. In traditional three-dimensional tunnel geological modeling, each step is often executed linearly and fragmented, lacking feedback adjustment mechanism, which leads to the fact that once there are parameter abnormal jumps, false strong and weak zones and other non-physical phenomena in the early stage of modeling, it is often difficult to be identified and corrected in time. This step updates the original lithology spatial variation scale index matrix by taking the backtracking analysis and abnormal identification results of the previous stage as the optimization entrance, improves the description accuracy of high variation areas, simultaneously updates the spatial unit division strategy to make the division scale more consistent with the new geological reality, and adjusts the parameter transmission constraint logic to ensure that the propagation path is strictly controlled by the reliable level and lithology continuity. Through this multi-dimensional linkage optimization process, not only the re-matching between the internal structure of the model and the external geological data is realized, but also the modeling process is transformed from "passive response" to "active correction". Finally, a closed-loop modeling logic chain composed of six links of lithology scale identification, unit division, grade control, propagation limitation, abnormal detection and feedback optimization is established, so that the model can continuously eliminate errors and enhance adaptability in each iteration, significantly improve the accuracy of geological structure expression, the rationality of rock mass parameter distribution and the reliability of engineering decision-making, and ensure that the modeling results have higher scientific value and engineering guidance significance.

[0057] The application realizes the transition from "geometric driving" to "geological driving" of spatial division by introducing a lithology spatial variation scale index matrix at the beginning of modeling; meanwhile, the introduction of credible level control and boundary propagation restriction in the parameter propagation process ensures that the interpolation process is subject to the dual constraints of structural continuity and data integrity, preventing the mispropagation of parameters in the lithology discontinuous area; further, through the multi-scale inversion verification and backtracking correction mechanism, the dynamic identification and local optimization of model errors are realized. Ultimately, the method establishes a full-process modeling system covering lithology identification, structure division, credible control, propagation restriction, anomaly detection and closed-loop optimization, significantly improving the expression accuracy and engineering adaptability of the model to complex geological structures, providing a more reliable geological basis for tunnel construction design and risk control, and having significant practical value and engineering promotion prospects.

[0058] The above formulas are dimensionless numerical calculations, and the formulas are obtained by software simulation of a large amount of data to obtain a formula of the nearest true situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0059] The above only describes some exemplary embodiments of the application by way of illustration, and it is needless to say that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the application. Therefore, the above drawings and descriptions are illustrative in nature and should not be understood as limiting the scope of protection of the claims of the application.

[0060] It should be noted that in this paper, if there are relationship terms such as first and second, they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the sentence "including a…" does not exclude the presence of other identical elements in the process, method, article or equipment including the element.

[0061] It should be understood that in various embodiments of the present application, the size of the sequence number of each process does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0062] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0063] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0064] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e. they can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiment according to actual needs.

[0065] In addition, the functional units in each embodiment of the present application can be integrated into one processing unit, or each unit can be physically present separately, or two or more units can be integrated into one unit.

[0066] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0067] The above only describes some exemplary embodiments of the present application by way of illustration, and it is self-evident that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present application. Therefore, the above figures and descriptions are illustrative in nature and should not be understood as limiting the scope of protection of the claims of the present application.

Claims

1. A tunnel geological modeling method, characterized in that, Includes the following steps: S1: Collect borehole data and lithological spatial distribution information, calculate the frequency of lithological changes in the horizontal and vertical directions and the fluctuation range of rock mass mechanical parameters of each rock layer, and construct a lithological spatial variation scale index matrix to characterize the spatial heterogeneity of the rock layers. S2, based on the lithological spatial variation scale index matrix, the modeling area is adaptively divided into spatial units to ensure that the division scale is consistent with the lithological variation scale, ensuring that the lithology within the unit is consistent and that the boundary avoids the location of lithological abrupt change, thus forming a spatial unit structure consistent with the lithological distribution. S3, based on the spatial unit structure, evaluates the structural integrity and borehole data coverage of the unit, determines the confidence level of each spatial unit, and sets parameter transmission constraints according to the confidence level to limit the propagation of parameters between spatial units that meet the confidence level requirements; S4. Based on parameter transfer constraints, boundary control is applied to parameter propagation between spatial units. A propagation threshold is set at the boundary to block parameter transfer between lithological discontinuities, ensuring that parameters only propagate within spatial units with similar lithology and acceptable confidence levels. S5, based on the completion of boundary propagation control, construct spatial unit parameter fields, carry out multi-scale inversion verification, identify non-physical parameter abrupt change areas with obvious parameter gradient changes or lithological transition areas, and conduct retrospective analysis and structural correction of anomalous areas; S6, based on the results of retrospective analysis and structural correction, collaboratively optimizes the lithological spatial variation scale index matrix, spatial unit division strategy, and parameter transfer constraints to form a closed-loop modeling process, thereby controlling structural modeling errors and improving the accuracy and engineering applicability of the tunnel's three-dimensional geological model.

2. The tunnel geological modeling method according to claim 1, characterized in that, Step S1 includes: Collect borehole data and establish a three-dimensional spatial database to extract the locations of the top and bottom interfaces of each rock stratum; Layered linear fitting was used to obtain the extension trend of rock strata, and local interpolation method was used to complete the interface. Calculate the frequency of lithological changes and the fluctuation range of rock mass mechanical parameters to form a two-dimensional variation index table; A lithological spatial variation scale index matrix is ​​formed by normalization and classification, and the basis for spatial unit division is determined by combining variation field visualization technology.

3. The tunnel geological modeling method according to claim 1, characterized in that, Step S2 includes: A spatial division requirement map was established based on the lithological spatial variation scale index matrix, dividing the area into a high variation sensitive zone, a medium variation transition zone, and a low variation stable zone. Implement a spatial unit division scale matching algorithm and adopt different division scales according to the level of variation; The weighted sliding scale window strategy is applied to adjust the position of the unit boundary so that it avoids lithological abrupt change zones; After the division is completed, consistency checks and borehole data coverage verification are carried out to ensure that the spatial structure and lithological distribution are consistent.

4. The tunnel geological modeling method according to claim 1, characterized in that, Step S3 includes: Assess the structural integrity of spatial units and identify the presence of fault crossings, lithological abrupt changes, or structural abnormalities. Calculate the coverage of borehole data and analyze the number of boreholes, distribution uniformity, and parameter completeness; The structural integrity score and the data coverage score are weighted and combined to classify them into high confidence level, medium confidence level and low confidence level; Based on the trust level, parameter transmission constraints are set, and a parameter propagation path control mechanism is constructed.

5. The tunnel geological modeling method according to claim 1, characterized in that, Step S4 includes: Identify the lithological property relationships and confidence level differences at the boundaries of spatial units, and establish a boundary relationship matrix; Based on the boundary relationship matrix, propagation threshold rules are set to distinguish between open boundaries, semi-open boundaries, and closed boundaries; A spatial propagation path tracing algorithm is used to monitor the interpolation path in real time and determine whether the propagation threshold condition is met. For paths that do not meet the propagation conditions, implement propagation blocking or interpolation correction to ensure that the propagation process is controlled by boundary restrictions.

6. The tunnel geological modeling method according to claim 1, characterized in that, Step S5 includes: Construct spatial unit parameter fields based on parameter interpolation results under boundary propagation control; The modeling space is divided into multiple scale segments, parameter gradient changes are calculated, and abnormal response regions are identified. Perform backtracking analysis on the abnormal response area to pinpoint the source of error; Based on the backtracking results, structural corrections are implemented, and the parameter field is reconstructed to verify whether the anomalies have been effectively eliminated.

7. The tunnel geological modeling method according to claim 1, characterized in that, Step S6 includes: The lithological spatial variation scale index matrix was revised based on the results of the retrospective analysis; The spatial unit partitioning strategy was optimized and adjusted based on the revised index matrix. Based on the structural adjustment results, the parameter transfer constraints between spatial units are reset; The parameter field is reconstructed and verified for evaluation, forming a closed-loop modeling process.

Images