Geotechnical engineering simulation test system
By constructing a geological twin model and dividing the steady-state and sensitive areas, and applying targeted testing factors, the problem of improper resource allocation in existing simulation testing systems has been solved, achieving efficient and targeted simulation testing for geotechnical engineering.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BUILDING MATERIALS GUANGZHOU ENG SURVEY INST CO
- Filing Date
- 2026-01-28
- Publication Date
- 2026-04-10
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing simulation testing systems cannot accurately identify and intelligently partition different regions based on the inherent stability differences of geological structures. This results in testing resources not being allocated as needed, making it difficult to conduct in-depth assessments of high-risk geological areas. Furthermore, they consume too many resources in areas with high stability and lack specificity.
Multidimensional geological data is acquired through the data acquisition module, a three-dimensional geological model is constructed and dynamically coupled into a geological twin model, steady-state and sensitive areas are divided, different test factors are applied to conduct simulation tests, the model is used to predict the simulation state and generate a test report.
This approach enables a deeper assessment of the safety and reliability of sensitive areas in geotechnical engineering, avoids excessive resource consumption in stable areas, and significantly improves the efficiency and relevance of simulation tests.
Smart Images

Figure CN121831104A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of engineering test technology, more particularly, the present application relates to a geotechnical engineering simulation test system. BACKGROUND
[0002] With the development of computer technology and artificial intelligence technology, simulation has become a core tool for geotechnical engineering analysis and judgment. By establishing a geomechanics model and simulating the application of virtual load or environmental change factors, the specific performance of the geotechnical engineering structure in stress, deformation and seepage and other multidimensional aspects can be accurately simulated and tested, so as to determine the overall safety and reliability of the geotechnical engineering.
[0003] The patent application with publication number CN110441497A discloses a deep geotechnical in-situ testing robot and its testing method. The robot can effectively improve the accuracy of test results without disturbing the original geotechnical body. It can perform static cone penetration testing in different geotechnical bodies, at any depth, and in drill holes with different temperatures and water contents, overcoming the limitations of existing testing methods. It can provide accurate and reliable basic physical and mechanical parameters for engineering design, model testing and numerical simulation of geotechnical engineering, water conservancy engineering, bridge and tunnel engineering, and mining engineering, and has good application prospects. The existing simulation test system usually adopts a unified test strategy when simulating and testing geotechnical engineering. It cannot accurately identify and intelligently partition different regions according to the internal stability differences of the geological structure, resulting in that the test resources cannot be allocated as needed in regions with different risk levels, making it difficult to conduct in-depth and focused evaluation and analysis of high-risk geological regions, and easily consuming too many analysis resources in geological regions with high stability, thus causing the simulation test process to lack pertinence and unable to effectively balance evaluation depth and test efficiency, failing to achieve the simulation test effect of teaching students according to their aptitude and divide and conquer.
[0004] In view of the above problems, the present application provides a geotechnical engineering simulation test system. SUMMARY
[0005] In order to overcome the above-mentioned defects of the prior art and achieve the above-mentioned purposes, the present application provides the following technical scheme: a geotechnical engineering simulation test system, comprising: A data acquisition module is used to expand the fixed profile of the geotechnical engineering outwardly into a test area and acquire multi-dimensional geological data of the test area, including geometric shape data, lithology distribution data, physical and mechanical data, and hydrological environment data. A model coupling module is used to construct a geological three-dimensional model through the multi-dimensional geological data and dynamically couple the geological three-dimensional model with the test area into a geological twin model by using a physical coupling engine. The simulation testing module is used to divide the geological twin model into a steady-state region and a sensitive region, and then apply the corresponding test factors to the steady-state region and the sensitive region respectively to conduct simulation tests. The model prediction module is used to extract regional test parameters from the steady-state region and the sensitive region. The regional test parameters include absolute displacement value, elastic-plastic quantity, local density value and pore water flow rate. The corresponding simulated state is predicted by the regional geological prediction model. The test analysis module is used to merge and stitch together the simulated states of the steady-state region and the sensitive region, determine the simulation test level of the geological twin model, and generate a simulation test report for geotechnical engineering.
[0006] Furthermore, the method for generating the test area is as follows: Using a one-year analysis period, the settlement amplitude of the construction site in the previous analysis period is retrieved. The difference between the settlement amplitude and the standard settlement value is then compared with the standard settlement value to calculate the oversettlement ratio. In the previous analysis period, mark A sampling points with interval distribution and query the soil density and soil moisture of the construction site at A sampling points; Add up the densities of A soils and calculate the average density. Then, calculate the difference between the average density and the standard density value, and compare it with the standard density value to calculate the excess density ratio. The average humidity is calculated by adding up the humidity of A soil samples. The difference between the average humidity and the standard humidity value is then compared with the standard humidity value to calculate the excess humidity ratio. The maximum depth value of geotechnical engineering is found by looking up the engineering design drawings, and the maximum depth value is combined with the oversettlement ratio, overdensity ratio and overwetness ratio to calculate the expansion factor; Locate the design boundary of the geotechnical engineering project on the construction site and obtain a fixed profile. Using the fixed profile as a reference, expand the fixed profile outward by a multiple to generate the expanded boundary. The construction site located inside the expanded boundary is recorded as the test area.
[0007] Furthermore, the method for constructing a three-dimensional geological model is as follows: The multidimensional geological data is filtered and denoised, and the filtered and denoised multidimensional geological data is then imported into the same coordinate system for coordinate alignment. Elevation points at the same cross section are extracted from lithological distribution data. Using the elevation points as a reference, the gradient direction, average value, and formation thickness of the top plate surface are set. The top plate surface of the cross section is generated by discrete smooth interpolation. All the top plate surfaces are then superimposed and spliced to generate the cross section volume. The dimensions of the cross-section are stretched along the X-axis, Y-axis, and Z-axis respectively until the outer contour of the cross-section is consistent with the geometric data, thus converting the cross-section into a cross-section model. The model database indexes soil and rock contours and hydrological contours that are compatible with physical and mechanical data and hydrological environment data. The hydrological contours are embedded into the soil and rock contours to form a hybrid contour. The hybrid contours are then imported into the cross-section model using rendering technology to construct a geological 3D model.
[0008] Furthermore, the method for constructing a geological twin model is as follows: The geological 3D model is divided into continuous triangular meshes using mesh generation technology, and all triangular meshes are assigned initial material parameters. Real-time acquisition of multidimensional geological data of the test area, marking of discretely distributed coupling points on the geological three-dimensional model, and determination of the mapping relationship between multidimensional geological data and coupling points one by one; Couple the multidimensional geological data with a positive mapping relationship to the coupling points to generate coupling units, and remove the coupling points that are not coupled. Multidimensional geological data and coupling points in all coupling units are synchronously input into the physical coupling engine, which enables the dynamic coupling of multidimensional geological data with the geological 3D model, and maps out a geological twin model.
[0009] Furthermore, the method for dividing the steady-state region and the sensitive region is as follows: Divide the previous analysis period into B consecutive sub-segments, measure the maximum length and maximum width of the fault where the C triangular grids are located one by one, measure the coverage area of the fault where the C triangular grids are located along the top view angle, add the maximum length and maximum width together and calculate the average, compare it with the coverage area of the fault, and calculate the size factor. Using the horizontal direction as the reference plane, the dip angle of the fault where the C triangular grids are located is measured one by one to obtain C dip values; Size factors greater than the calibrated size threshold are recorded as steady-state parameters, tilt values less than the calibrated tilt threshold are recorded as steady-state parameters, triangular meshes with a number of 2 steady-state parameters are recorded as steady-state meshes, and the remaining triangular meshes are recorded as sensitive meshes. Based on the standard that there is no overlap between two adjacent triangular grids, the steady-state grid and the sensitive grid are extended into steady-state region and sensitive region in the geological twin model.
[0010] Furthermore, the testing factors include meteorological and hydrological factors, active load factors, geophysical factors, and chemical erosion factors; Meteorological and hydrological factors include precipitation sub-factors, extreme cold sub-factors, and soil erosion sub-factors; Active load factors include load sub-factors and excavation sub-factors; Geophysical factors include crustal movement and soil permeability. Chemical corrosion factors include pH factors, microbial factors, and heavy metal factors.
[0011] Furthermore, the simulation testing methods for the steady-state region and the sensitive region are as follows: A1: Randomly select one sub-factor from meteorological and hydrological factors, active load factors, geological and physical factors, and chemical erosion factors to obtain four sub-factors; A2: Simulation scenarios adapted to the four sub-factors are constructed using simulation software, and the four simulation scenarios are numbered in ascending order. A3: The four simulated scenarios are merged and superimposed into a hybrid scenario, and the hybrid scenario is applied to D steady-state regions in sequence to conduct simulation tests on the D steady-state regions; A4: Apply the four simulated scenarios to the E sensitive areas in ascending order of their numbers, and conduct simulated tests on the E sensitive areas. A5: Repeat steps A1-A5 until all sub-factors among meteorological and hydrological factors, active load factors, geophysical factors, and chemical erosion factors have been selected and simulated, then stop.
[0012] Furthermore, the simulated states include unbalanced states and normal states; The method for predicting the simulated state is as follows: Multiple sets of regional test parameters are pre-collected in the steady-state region and the sensitive region under both imbalanced and normal states. Each set of regional test parameters is labeled as a training feature. The simulated state of each set of training features is labeled, and the imbalanced state is converted to 0 and the normal state is converted to 1. The labeled training features are divided into a training set and a test set. The regional geological prediction model is trained using the training set and tested using the test set. When the mean of the prediction error of all training features in the test set is less than the error threshold, the regional geological prediction model is output. When the output of the regional geological prediction model is 0, the simulation state is unbalanced; when the output of the regional geological prediction model is 1, the simulation state is normal.
[0013] Furthermore, the simulation test levels include a safety and reliability level, a slight imbalance level, and a runaway risk level; The steady-state region and sensitive region that are in an unbalanced state in the simulation are recorded as target regions, and all target regions are marked one by one in the geological twin model; The target areas in adjacent positions are combined into a splicing area, the volume of the splicing area is calculated and recorded as the splicing volume, and the splicing volume is compared with the total volume of the geological twin model to calculate the splicing ratio. The volume of each target area was measured one by one, and the accumulated volume of the areas was compared with the total volume of the geological twin model to calculate the imbalance ratio. When the splicing ratio is greater than the calibrated splicing threshold, the splicing ratio is recorded as the out-of-control parameter; when the imbalance ratio is greater than the calibrated imbalance threshold, the imbalance ratio is recorded as the out-of-control parameter. If the number of runaway parameters in the geological twin model is 2, 1, and 0 respectively, the simulation test levels are runaway danger level, mild imbalance level, and safe and reliable level respectively.
[0014] Furthermore, the simulation test report includes the simulation status of the steady-state region, the simulation status of the sensitive region, the spatial location of the target region, the number of target regions, the spatial location of the splicing region, and the simulation test level.
[0015] The technical advantages of the geotechnical engineering simulation testing system of the present invention are as follows: This invention, through the analysis and calculation of structural indices such as fault size and tilt angle in geological twin models, can intelligently divide geological twin models into two forms: stable and sensitive regions. By applying different test factors tailored to the differences in mechanical properties and risk levels between the stable and sensitive regions, the analytical resources for simulation testing can be focused on the higher-risk sensitive regions. This not only enhances the depth of safety and reliability assessment in sensitive regions of geotechnical engineering but also avoids excessive resource consumption in stable regions, thus significantly improving the efficiency and relevance of overall simulation testing. It achieves a differentiated and targeted simulation testing effect in geotechnical engineering. Attached Figure Description
[0016] Figure 1 This is a schematic diagram of a geotechnical engineering simulation testing system provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart illustrating a geotechnical engineering simulation testing method provided in Embodiment 2 of the present invention. Detailed Implementation
[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0018] Example 1: Please refer to Figure 1 As shown in this embodiment, a geotechnical engineering simulation testing system includes: The data acquisition module determines the fixed outline of the geotechnical engineering, expands the fixed outline outward to form the test area, and collects multidimensional geological data of the test area. In actual construction, geotechnical engineering extends into the interior of the rock and soil in the construction area. Although geotechnical engineering has static closed boundaries on the design drawings, the interior of the rock and soil is dynamically interconnected, which means that there will be mutual influence between adjacent rock and soil. Therefore, it is necessary to identify and determine the areas actually involved and affected by geotechnical engineering. In this embodiment, the fixed profile refers to the boundary line formed by the location directly involved in the construction of geotechnical engineering within the rock and soil, which serves as the initial boundary for subsequent simulation tests of geotechnical engineering.
[0019] After determining the fixed contour, it is necessary to expand the fixed contour outward, and the area obtained after expansion and diffusion is recorded as the test area of this simulation test; Specifically, the method for generating the test area is as follows: The design boundaries of geotechnical engineering are found by looking up the engineering design drawings, and the location of the design boundaries is located on the construction site to obtain a fixed outline; Using a one-year analysis period, the settlement amplitude of the construction site in the previous analysis period is retrieved. The difference between the settlement amplitude and the standard settlement value is then compared with the standard settlement value to calculate the oversettlement ratio. The standard settlement value refers to the maximum settlement amplitude of the construction site within a pre-set analysis period. The formula for calculating the oversinking ratio is: ; In the formula, The ratio of underweight to overweight. For the settlement range, This is the standard settlement value; In the previous analysis period, mark A sampling points with interval distribution and query the soil density and soil moisture of the construction site at A sampling points; The average density is calculated by adding up the densities of A soil samples. The difference between the average density and the standard density value is then compared with the standard density value to calculate the excess density ratio. The standard density value refers to the maximum soil density of the pre-set construction site within one analysis period. The formula for calculating the ultra-dense ratio is: ; In the formula, The ultra-dense ratio, For average density, This is the standard density value; The average humidity is calculated by adding up the humidity of A soil samples. The difference between the average humidity and the standard humidity value is then compared with the standard humidity value to calculate the excess humidity ratio. The standard humidity value refers to the maximum soil humidity of the pre-set construction site within one analysis period. The formula for calculating the excess humidity ratio is: ; In the formula, This is the humidity ratio. The average humidity. This is the standard humidity value; The maximum depth value of geotechnical engineering is found by looking up the engineering design drawings, and the maximum depth value is combined with the oversettlement ratio, overdensity ratio and overwetness ratio to calculate the expansion factor; The formula for calculating the expansion factor is: ; In the formula, To increase the number of columns, This is the maximum depth value; Using a fixed profile as a reference, the fixed profile is expanded outward by a factor of one expansion to generate an expansion boundary, and the construction site located inside the expansion boundary is recorded as the test area.
[0020] It should be noted that the relative positions of the test areas in different analysis cycles are not static. The positions of the test areas change with factors such as maximum depth, settlement, density, and humidity in the construction site. In this embodiment, the test areas are determined only based on the relevant factors analysis of the previous analysis cycle.
[0021] Multidimensional geological data is used to specifically represent the multidimensional performance parameters of rocks and soils in the test area, serving as a comprehensive representation of the physical properties of rocks and soils in the test area. Specifically, multidimensional geological data includes, but is not limited to, geometric morphology data, lithological distribution data, physical and mechanical data, and hydrological and environmental data; Geometric morphological data is used to represent the geometric dimensions and morphological distribution of rocks and soil in the test area, providing morphological dimension data support for the modeling of the test area.
[0022] Lithological distribution data is used to represent the fault strike and spatial distribution of rocks and soils in the test area, providing lithological dimension data support for the modeling of the test area.
[0023] Physical and mechanical data are used to represent the physical properties and mechanical characteristics of rocks and soil in the test area, providing mechanical dimension data support for the modeling of the test area.
[0024] Hydrological and environmental data are used to represent the hydrological fluctuations and environmental changes in the rocks and soil of the test area, providing environmental data support for the modeling of the test area.
[0025] In this embodiment, when collecting multidimensional geological data, it is necessary to select the corresponding acquisition equipment and acquisition method according to the different multidimensional geological data. For example, when collecting geometric morphology data, it is obtained through borehole exploration equipment and remote sensing equipment, and when collecting hydrological environment data, it is obtained through meteorological equipment.
[0026] The model coupling module constructs a three-dimensional geological model using multi-dimensional geological data, and dynamically couples the three-dimensional geological model with the test area through a physical coupling engine to construct a geological twin model. A geological 3D model is a 3D model constructed using 3D modeling technology based on multi-dimensional geological data, which is consistent with the geological morphology and geometric structure of the test area. It can convert the physical test area into a virtual simulation model, which is convenient for simulating testing operations in geotechnical engineering. The method for constructing a three-dimensional geological model is as follows: The multidimensional geological data is filtered and denoised, and the filtered and denoised multidimensional geological data is then imported into the same coordinate system for coordinate alignment. Elevation points within the same cross section are extracted from lithological distribution data. Using these elevation points as a reference, the gradient direction, average value, and formation thickness of the top plate surface are set. The top plate surface of the cross section is generated using a discrete smooth interpolation method. All the top plate surfaces are then sequentially superimposed and stitched together to generate the cross section volume. The top plate surface is a model surface generated based on the cross section and elevation points, serving as a component of the cross section volume. The discrete smooth interpolation method is existing technology in this field and will not be described in detail here. The dimensions of the cross-section are stretched along the X-axis, Y-axis, and Z-axis respectively until the outer contour of the cross-section is consistent with the geometric data, thus converting the cross-section into a cross-section model. The model database indexes soil and rock contours and hydrological contours that are compatible with physical and mechanical data and hydrological environment data. The hydrological contours are embedded into the soil and rock contours to form a hybrid contour. The hybrid contours are then imported into the cross-section model using rendering technology to construct a geological 3D model.
[0027] The constructed geological 3D model can only represent the actual situation of the test area in the previous analysis cycle, but it cannot be mapped and associated with the test area in real time and dynamically. This makes the data information on the geological 3D model relatively static and simple, and has limitations. For the reasons mentioned above, it is necessary to dynamically couple the geological 3D model with the test area, so as to map the real-time changes of the test area onto the geological 3D model, thereby converting the geological 3D model into a geological twin model that is mapped to the test area in real time.
[0028] The physical coupling engine is used to couple and map the real-time changes in the physical information of the test area onto the geological 3D model, enabling real-time data interaction between two different entities. The physical coupling engine does not directly use the coupler or coupling algorithm in the existing technology, but is obtained by optimizing and upgrading the coupler or coupling algorithm, so that the physical coupling engine can couple and map the real-time changes in the test area onto the geological 3D model.
[0029] Specifically, the method for constructing a geological twin model is as follows: The geological 3D model is divided into continuous triangular meshes using mesh generation technology, and all triangular meshes are assigned initial material parameters (e.g., elastic modulus, Poisson's ratio, permeability coefficient, etc.). This ensures that the triangular meshes remain independent in subsequent region division, improving the accuracy of region division. Multidimensional geological data of the test area is collected in real time, discrete coupling points are marked on the geological 3D model, and the mapping relationship between multidimensional geological data and coupling points is determined one by one. The mapping relationship is used to specifically represent the mapping result between multidimensional geological data and coupling points. The mapping relationship includes forward mapping and reverse mapping. They are used to refer to multidimensional geological data and coupling points that can be coupled and cannot be coupled and bound, respectively. Couple the multidimensional geological data with a positive mapping relationship to the coupling points to generate coupling units, and remove the coupling points that are not coupled. Multidimensional geological data and coupling points in all coupling units are synchronously input into the physical coupling engine, which enables the dynamic coupling of multidimensional geological data with the geological 3D model, and maps out a geological twin model.
[0030] It should be noted that by using synchronous input dynamic coupling, the consistency of coupling between multidimensional geological data and three-dimensional geological models can be ensured, preventing deviations in coupling results at different times during dynamic coupling.
[0031] The simulation test module divides the geological twin model into steady-state and sensitive regions, applies corresponding test factors to the steady-state and sensitive regions respectively, and performs simulation tests on the steady-state and sensitive regions. Geological twin models are used to map and represent the real-time changes in a test area as a whole. This means that geological twin models can only simulate and test from a macroscopic level and cannot perform targeted simulation and testing of specific local areas within the test area. To enable targeted simulation testing of specific locations within the test area, the geological twin model of the overall structure needs to be divided into regions. This allows for the breaking down of the geological twin model into smaller parts, achieving targeted simulation testing of specific local areas.
[0032] After dividing the geological twin model, two types of regions can be formed: a stable region and a sensitive region; and the geological twin model only contains the stable region and the sensitive region.
[0033] In this embodiment, the steady-state region refers to the region in the geological twin model where the structural morphology and geological properties are relatively stable and not easily changed; the sensitive region refers to the region in the geological twin model where the structural morphology and geological properties are relatively fluctuating and easily changed.
[0034] The method for dividing the steady-state region and the sensitive region is as follows: Divide the previous analysis period into B sub-segments, and make the last moment of the previous sub-segment and the first moment of the next sub-segment adjacent moments; The maximum length and maximum width of the fault where each of the C triangular grids is located are measured one by one. The coverage area of the fault where the C triangular grids are located is measured along the top view angle. The maximum length and maximum width are added together and averaged. The average is then compared with the coverage area of the fault to calculate the size factor. The formula for calculating the size factor is: ; In the formula, This is a size factor. For the maximum length, For the maximum width, The area covered by the fault; Using the horizontal direction as the reference plane, the dip angle of the fault where the C triangular grids are located is measured one by one to obtain C dip values; Size factors greater than the calibrated size threshold are recorded as steady-state parameters, and tilt values less than the calibrated tilt threshold are recorded as steady-state parameters. The number of steady-state parameters in the triangular mesh is counted, and triangular meshes with 2 steady-state parameters are recorded as steady-state meshes, while the remaining triangular meshes are recorded as sensitive meshes. The calibrated size threshold and calibrated tilt threshold refer to the minimum and maximum values of the size factor and tilt value when recorded as steady-state parameters, respectively, which ensures the accuracy of steady-state parameter identification. Based on the standard that there is no overlap between two adjacent triangular grids, the steady-state grid and the sensitive grid are expanded within the geological twin model. The steady-state grid is expanded into a steady-state region, and the sensitive grid is expanded into a sensitive region, resulting in D steady-state regions and E sensitive regions.
[0035] It should be noted that the quantitative relationship between the steady-state region and the sensitive region is not limited, nor is the positional relationship between the two, which ensures that the steady-state region and the sensitive region can accurately and comprehensively divide the geological twin model.
[0036] After obtaining the steady-state region and the sensitive region, separate and targeted simulation tests can be carried out on the steady-state region and the sensitive region. Since the actual geological structure and performance of the steady-state region and the sensitive region in the test area are not the same, it is necessary to apply different test factors to the steady-state region and the sensitive region respectively in order to achieve the effect of individualized teaching. In this embodiment, the test factor is a simulation command applied to the geological twin model through virtual simulation technology to change the geological morphology and properties of the steady-state region and the sensitive region.
[0037] Specifically, the testing factors include meteorological and hydrological factors, active load factors, geophysical factors, and chemical erosion factors; Meteorological and hydrological factors refer to the simulation tests conducted on the geotechnical engineering locations corresponding to steady-state and sensitive areas based on natural environments such as meteorology and hydrology; specifically, meteorological and hydrological factors include, but are not limited to, rainfall sub-factors, extreme cold sub-factors, and soil erosion sub-factors.
[0038] Active load factors refer to the simulation tests conducted on the geotechnical engineering locations corresponding to steady-state and sensitive areas from the perspective of human activities and mechanical movements; specifically, active load factors include, but are not limited to, load sub-factors and excavation sub-factors.
[0039] Geophysical factors refer to the simulation testing of geotechnical engineering locations corresponding to stable and sensitive areas from geological levels such as geographical structure and morphology; specifically, geophysical factors include, but are not limited to, crustal movement sub-factors and soil permeability sub-factors.
[0040] Chemical erosion factors refer to the simulation tests conducted on the geotechnical engineering locations corresponding to stable and sensitive areas from the chemical level, such as chemical erosion. Specifically, chemical erosion factors include, but are not limited to, pH factors, microbial factors, and heavy metal factors.
[0041] When conducting simulation tests on the steady-state region and the sensitive region, it is necessary to apply corresponding test factors to the steady-state region and the sensitive region to ensure that the steady-state region and the sensitive region can be accurately simulated and tested. Specifically, the simulation test methods for the steady-state region and the sensitive region are as follows: A1: Randomly select one sub-factor from meteorological and hydrological factors, active load factors, geological and physical factors, and chemical erosion factors to obtain four sub-factors; A2: Simulation scenarios adapted to the four sub-factors are constructed using simulation software, and the four simulation scenarios are numbered in ascending order. The simulation scenarios serve as the test environment for simulating the steady-state region and the sensitive region, thereby ensuring the accuracy of the simulation test operations at the virtual level. A3: The four simulated scenarios are merged and superimposed into a hybrid scenario, and the hybrid scenario is applied to D steady-state regions in sequence to conduct simulation tests on the D steady-state regions; A4: Apply the four simulated scenarios to the E sensitive areas in ascending order of their numbers, and conduct simulated tests on the E sensitive areas. A5: Repeat steps A1-A5 until all sub-factors among meteorological and hydrological factors, active load factors, geophysical factors, and chemical erosion factors have been selected and simulated, then stop.
[0042] It should be noted that by mixing and applying simulation scenarios individually, targeted simulation testing operations can be performed on steady-state and sensitive regions, thus achieving the effect of on-demand allocation and simulation testing of steady-state and sensitive regions.
[0043] The model prediction module extracts regional test parameters from the steady-state region and the sensitive region, and predicts the corresponding simulated state through the regional geological prediction model; After conducting simulation tests on the steady-state region and the sensitive region, both the steady-state region and the sensitive region will experience certain changes and fluctuations. These changes and fluctuations can be used to analyze and judge the actual performance of each region in the simulation test. Regional test parameters are parameters used to represent the changes in steady-state and sensitive regions after simulation testing; Specifically, the regional test parameters include absolute displacement value, elastic-plasticity value, local density value, and pore water flow rate; The absolute displacement value refers to the absolute distance that the soil and rock position in the steady-state region and the sensitive region moves from before the simulation test to after the simulation test after the application of test factors, and is used as the result of the deformation field simulation test. In this embodiment, the absolute displacement value is obtained by measuring the spatial coordinates of the anchor points in the steady-state region and the sensitive region before and after the simulation test.
[0044] Elastic plasticity refers to the cumulative amount of irreversible plastic deformation that occurs in the soil and rock locations in the steady-state and sensitive regions after the application of test factors, from before the simulation test to after the simulation test, as a result of strain field simulation test; In this embodiment, the elastic plasticity is obtained by measuring and calculating the degree of volumetric deformation in the steady-state region and the sensitive region before and after the simulation test.
[0045] Local density value refers to the magnitude of change in density of rock and soil in steady-state and sensitive areas from before to after the simulation test after the application of test factors, serving as another result of strain field simulation test; In this embodiment, the local density value is obtained by measuring the compactness of rock and soil per unit volume in the steady-state region and the sensitive region before and after the simulation test.
[0046] Pore water flow rate refers to the degree of change in the water flow rate in the pores of the soil and rock in the steady-state region and the sensitive region after the application of test factors, from before the simulation test to after the simulation test, and is used as the result of the seepage field simulation test; In this embodiment, the pore water flow rate is obtained by measuring the difference in pore water flow rate between the steady-state region and the sensitive region before and after the simulation test.
[0047] After extracting the regional test parameters of the steady-state region and the sensitive region, the regional test parameters can be input into the regional geological prediction model to predict the corresponding simulation state, so that the simulation state can represent the specific results of the simulation test of the steady-state region and the sensitive region. Specifically, the simulated states include unbalanced states and normal states.
[0048] The regional geological prediction model is a model trained by combining a large number of regional test parameters and corresponding simulation states based on machine learning technology. It can intelligently predict and analyze the specific results of simulation tests in steady-state areas and sensitive areas. Specifically, the method for predicting the simulated state is as follows: Multiple sets of test parameters were pre-collected for the steady-state region and the sensitive region under both unbalanced and normal conditions. Each set of regional test parameters is labeled as a training feature, and the simulated state of each set of training features is labeled, including imbalanced state and normal state. The imbalanced state and normal state are converted into numerical labels respectively. For example, the imbalanced state is converted to 0 and the normal state is converted to 1. The labeled training features are divided into a training set and a test set; 70% of the training features are used as the training set and 30% of the training features are used as the test set; the regional geological prediction model is trained using the training set and tested using the test set. A preset error threshold is set. When the mean of the prediction errors of all training features in the test set is less than the error threshold, the regional geological prediction model is output. The regional test parameters of the steady-state region and the sensitive region are respectively input into the regional geological prediction model for prediction. When the output of the regional geological prediction model is 0, the simulation state is unbalanced; when the output of the regional geological prediction model is 1, the simulation state is normal.
[0049] In this embodiment, the regional geological prediction model can be either a support vector machine model or a random forest model; the preset error threshold is set in advance according to the actual accuracy required by the regional geological prediction model.
[0050] The test analysis module merges and stitches together the simulated states of the steady-state region and the sensitive region to determine the simulation test level of the geological twin model and generate a simulation test report for geotechnical engineering. The simulated states of the predicted steady-state and sensitive regions are discrete and cannot directly represent the overall geological twin model. Therefore, it is necessary to fuse and stitch the discrete simulated states to form a simulation test level that can represent the overall simulation test results of the geological twin model. In this embodiment, the simulation test levels include a safety and reliability level, a slight imbalance level, and a runaway risk level; the safety levels corresponding to the safety and reliability level, the slight imbalance level, and the runaway risk level are from high to low.
[0051] The method for determining the simulation test level is as follows: The steady-state region and sensitive region that are in an unbalanced state in the simulation are recorded as target regions, and all target regions are marked one by one in the geological twin model; The target areas in adjacent positions are combined into a splicing area, the volume of the splicing area is calculated and recorded as the splicing volume, and the splicing volume is compared with the total volume of the geological twin model to calculate the splicing ratio. The volume of each target area was measured one by one, and the accumulated volume of the areas was compared with the total volume of the geological twin model to calculate the imbalance ratio. When the splicing ratio exceeds the calibrated splicing threshold, it indicates that the volume of the spliced area has reached a relatively serious level, and the splicing ratio is recorded as an out-of-control parameter. The calibrated splicing threshold refers to the minimum value of the splicing ratio when it is recorded as an out-of-control parameter. When the imbalance ratio is greater than the calibrated imbalance threshold, it indicates that the volume of the target area has reached a relatively serious level, and the imbalance ratio is recorded as the out-of-control parameter; the calibrated imbalance threshold refers to the minimum value of the imbalance ratio when it is recorded as the out-of-control parameter. If the number of runaway parameters in the geological twin model is 2, it indicates that the safety of the simulation test in the test area is low, and the simulation test level is recorded as the runaway danger level. If the number of out-of-control parameters in the geological twin model is 1, it indicates that the safety of the simulation test in the test area is moderate, and the simulation test level is recorded as a mild imbalance level. If the number of out-of-control parameters in the geological twin model is 0, it indicates that the simulation test in the test area is highly safe, and the simulation test level is recorded as a safe and reliable level.
[0052] After determining the simulation test level of the geological twin model, the safety and stability performance of the geotechnical engineering corresponding to the test area can be intuitively represented. Based on the relevant simulation test results obtained above, a simulation test report of the geotechnical engineering can be formulated, thereby providing a comprehensive display of the simulation test results of the geotechnical engineering. In this embodiment, when generating the simulation test report, the report needs to include various different information, including but not limited to the simulation status of the steady-state region, the simulation status of the sensitive region, the spatial location of the target region, the number of target regions, the spatial location of the splicing region, and the simulation test level. This can record and summarize multi-dimensional information for geotechnical engineering throughout the entire simulation test process, facilitating the subsequent review and analysis of the simulation test results, and providing accurate, reasonable, and reliable theoretical support for subsequent construction operations in geotechnical engineering.
[0053] Example 2: Please refer to Figure 2 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A geotechnical engineering simulation testing method is provided, implemented based on a geotechnical engineering simulation testing system, including: S01: Expand the fixed contour of the geotechnical engineering project outward to form a test area, and collect multidimensional geological data of the test area; S02: Construct a three-dimensional geological model using multi-dimensional geological data, and use a physical coupling engine to dynamically couple the three-dimensional geological model with the test area to form a geological twin model; S03: Divide the geological twin model into a steady-state region and a sensitive region, and then apply the corresponding test factors to the steady-state region and the sensitive region respectively before conducting simulation tests; S04: Extract regional test parameters from the steady-state region and the sensitive region, and predict the corresponding simulated state through the regional geological prediction model; S05: Merge and stitch together the simulated states of the steady-state region and the sensitive region to determine the simulation test level of the geological twin model and generate a simulation test report for geotechnical engineering.
[0054] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. A geotechnical engineering simulation testing system, characterized in that, include: The data acquisition module is used to expand the fixed outline of the geotechnical engineering into a test area and collect multidimensional geological data of the test area, including geometric morphology data, lithological distribution data, physical and mechanical data and hydrological environmental data. The model coupling module is used to construct a three-dimensional geological model from multi-dimensional geological data and use the physical coupling engine to dynamically couple the three-dimensional geological model with the test area to form a geological twin model. The simulation testing module is used to divide the geological twin model into a steady-state region and a sensitive region, and then apply the corresponding test factors to the steady-state region and the sensitive region respectively to conduct simulation tests. The model prediction module is used to extract regional test parameters from the steady-state region and the sensitive region. The regional test parameters include absolute displacement value, elastic plasticity, local density value and pore water flow rate. The corresponding simulated state is predicted by the regional geological prediction model. The test analysis module is used to merge and stitch together the simulated states of the steady-state region and the sensitive region, determine the simulation test level of the geological twin model, and generate a simulation test report for geotechnical engineering.
2. The geotechnical engineering simulation testing system according to claim 1, characterized in that, The test area is generated as follows: Using a one-year analysis period, the settlement amplitude of the construction site in the previous analysis period is retrieved. The difference between the settlement amplitude and the standard settlement value is then compared with the standard settlement value to calculate the oversettlement ratio. In the previous analysis period, mark A sampling points with interval distribution and query the soil density and soil moisture of the construction site at A sampling points; Add up the densities of A soils and calculate the average density. Then, calculate the difference between the average density and the standard density value, and compare it with the standard density value to calculate the excess density ratio. The average humidity is calculated by adding up the humidity of A soil samples. The difference between the average humidity and the standard humidity value is then compared with the standard humidity value to calculate the excess humidity ratio. The maximum depth value of geotechnical engineering is found by looking up the engineering design drawings, and the maximum depth value is combined with the oversettlement ratio, overdensity ratio and overwetness ratio to calculate the expansion factor; Locate the design boundary of the geotechnical engineering project on the construction site and obtain a fixed profile. Using the fixed profile as a reference, expand the fixed profile outward by a multiple to generate the expanded boundary. The construction site located inside the expanded boundary is recorded as the test area.
3. The geotechnical engineering simulation testing system according to claim 2, characterized in that, The method for constructing a three-dimensional geological model is as follows: The multidimensional geological data is filtered and denoised, and the filtered and denoised multidimensional geological data is then imported into the same coordinate system for coordinate alignment. Elevation points at the same cross section are extracted from lithological distribution data. Using the elevation points as a reference, the gradient direction, average value, and formation thickness of the top plate surface are set. The top plate surface of the cross section is generated by discrete smooth interpolation. All the top plate surfaces are then superimposed and spliced to generate the cross section volume. The dimensions of the cross-section are stretched along the X-axis, Y-axis, and Z-axis respectively until the outer contour of the cross-section is consistent with the geometric data, thus converting the cross-section into a cross-section model. The model database indexes soil and rock contours and hydrological contours that are compatible with physical and mechanical data and hydrological environment data. The hydrological contours are embedded into the soil and rock contours to form a hybrid contour. The hybrid contours are then imported into the cross-section model using rendering technology to construct a geological 3D model.
4. The geotechnical engineering simulation testing system according to claim 3, characterized in that, The method for constructing a geological twin model is as follows: The geological 3D model is divided into continuous triangular meshes using mesh generation technology, and all triangular meshes are assigned initial material parameters. Real-time acquisition of multidimensional geological data of the test area, marking of discretely distributed coupling points on the geological three-dimensional model, and determination of the mapping relationship between multidimensional geological data and coupling points one by one; Couple the multidimensional geological data with a positive mapping relationship to the coupling points to generate coupling units, and remove the coupling points that are not coupled. Multidimensional geological data and coupling points in all coupling units are synchronously input into the physical coupling engine, which enables the dynamic coupling of multidimensional geological data with the geological 3D model, and maps out a geological twin model.
5. The geotechnical engineering simulation testing system according to claim 4, characterized in that, The method for dividing the steady-state region and the sensitive region is as follows: Divide the previous analysis period into B consecutive sub-segments, measure the maximum length and maximum width of the fault where the C triangular grids are located one by one, measure the coverage area of the fault where the C triangular grids are located along the top view angle, add the maximum length and maximum width together and calculate the average, compare it with the coverage area of the fault, and calculate the size factor. Using the horizontal direction as the reference plane, the dip angle of the fault where the C triangular grids are located is measured one by one to obtain C dip values; Size factors greater than the calibrated size threshold are recorded as steady-state parameters, tilt values less than the calibrated tilt threshold are recorded as steady-state parameters, triangular meshes with a number of 2 steady-state parameters are recorded as steady-state meshes, and the remaining triangular meshes are recorded as sensitive meshes. Based on the standard that there is no overlap between two adjacent triangular grids, the steady-state grid and the sensitive grid are extended into steady-state region and sensitive region in the geological twin model.
6. The geotechnical engineering simulation testing system according to claim 5, characterized in that, The test factors include meteorological and hydrological factors, active load factors, geological and physical factors, and chemical erosion factors; Meteorological and hydrological factors include precipitation sub-factors, extreme cold sub-factors, and soil erosion sub-factors; Active load factors include load sub-factors and excavation sub-factors; Geophysical factors include crustal movement and soil permeability. Chemical corrosion factors include pH factors, microbial factors, and heavy metal factors.
7. The geotechnical engineering simulation testing system according to claim 6, characterized in that, The simulation test methods for the steady-state region and the sensitive region are as follows: A1: Randomly select one sub-factor from meteorological and hydrological factors, active load factors, geological and physical factors, and chemical erosion factors to obtain four sub-factors; A2: Simulation scenarios adapted to the four sub-factors are constructed using simulation software, and the four simulation scenarios are numbered in ascending order. A3: The four simulated scenarios are merged and superimposed into a hybrid scenario, and the hybrid scenario is applied to D steady-state regions in sequence to conduct simulation tests on the D steady-state regions; A4: Apply the four simulated scenarios to the E sensitive areas in ascending order of their numbers, and conduct simulated tests on the E sensitive areas. A5: Repeat steps A1-A5 until all sub-factors among meteorological and hydrological factors, active load factors, geophysical factors, and chemical erosion factors have been selected and simulated, then stop.
8. The geotechnical engineering simulation testing system according to claim 7, characterized in that, The simulated states include unbalanced states and normal states; The method for predicting the simulated state is as follows: Multiple sets of regional test parameters are pre-collected in steady-state and sensitive regions under imbalance and normal states. Each set of regional test parameters is marked as a training feature. The simulated state of each set of training features is labeled, and the imbalance state is converted to 0 and the normal state is converted to 1. The labeled training features are divided into a training set and a test set. The regional geological prediction model is trained using the training set and tested using the test set. When the mean of the prediction error of all training features in the test set is less than the error threshold, the regional geological prediction model is output. When the output of the regional geological prediction model is 0, the simulation state is unbalanced; when the output of the regional geological prediction model is 1, the simulation state is normal.
9. A geotechnical engineering simulation testing system according to claim 8, characterized in that, The simulation test levels include a safety and reliability level, a slight imbalance level, and a runaway risk level; The steady-state region and sensitive region that are in an unbalanced state in the simulation are recorded as target regions, and all target regions are marked one by one in the geological twin model; The target areas in adjacent positions are combined into a splicing area, the volume of the splicing area is calculated and recorded as the splicing volume, and the splicing volume is compared with the total volume of the geological twin model to calculate the splicing ratio. The volume of each target area was measured one by one, and the accumulated volume of the areas was compared with the total volume of the geological twin model to calculate the imbalance ratio. When the splicing ratio is greater than the calibrated splicing threshold, the splicing ratio is recorded as the out-of-control parameter; when the imbalance ratio is greater than the calibrated imbalance threshold, the imbalance ratio is recorded as the out-of-control parameter. If the number of runaway parameters in the geological twin model is 2, 1, and 0 respectively, the simulation test levels are runaway danger level, mild imbalance level, and safe and reliable level respectively.
10. A geotechnical engineering simulation testing system according to claim 9, characterized in that, The simulation test report includes the simulation status of the steady-state region, the simulation status of the sensitive region, the spatial location of the target region, the number of target regions, the spatial location of the splicing region, and the simulation test level.
Citation Information
Patent Citations
Deep rock-soil body in-situ testing robot and testing method thereof
CN110441497A