Construction method of three-dimensional geotechnical engineering investigation information model based on BIM

Abstract

The present invention relates to the field of geotechnical engineering digital technology, and specifically to a method for constructing a three-dimensional geotechnical engineering survey information model based on BIM. The steps of the construction method include: obtaining multi-dimensional geotechnical engineering survey data, collecting geotechnical engineering survey data in real time from multiple sources, including geological borehole location coordinates, soil sample properties, groundwater level dynamic values, and geological structure maps, and performing data synchronization operations during the collection process to ensure timestamp consistency. The present invention fundamentally breaks the technical bottleneck of static physical process in traditional modeling by synergizing the geotechnical information generation mechanism of multi-physical field coupling with the dynamic response architecture of the BIM framework.

Description

Technical Field

[0001] The present invention relates to the field of geotechnical engineering digital technology, and specifically to a method for constructing a three-dimensional geotechnical engineering survey information model based on BIM. Background Art

[0002] In the field of geotechnical engineering digitization, the construction of BIM-based survey information models faces a technical bottleneck: traditional methods rely on static geological interpolation algorithms and are unable to couple the dynamic thermal-mechanical interaction mechanisms within geotechnical structures. While existing technologies can integrate borehole data with the BIM spatial framework, they lack physically constrained mathematical representations of the property field evolution of heterogeneous geotechnical structures, such as the stress redistribution induced by seepage-temperature fields. Especially when faced with multi-source, heterogeneous survey data, such as dynamic groundwater levels and tectonic stress fields, conventional models ignore the inherent energy transfer laws of geological processes, resulting in an output model that fails to reflect the progressive degradation behavior of geotechnical structures under engineering disturbances. The root cause lies in the lack of a theoretical framework that unifies geomechanical principles, thermodynamic responses, and BIM spatial topology within a dynamic field reconstruction, resulting in frequent systematic biases in engineering risk assessments. Therefore, the technical problem addressed in this paper is how to construct a geotechnical information model capable of coupled thermal-mechanical field responses, thereby addressing the problem of engineering decision-making distortion caused by the static nature of physical mechanisms. Summary of the Invention

[0003] The present disclosure proposes a method for constructing a three-dimensional geotechnical engineering survey information model based on BIM, aiming to overcome at least one defect existing in the prior art.

[0004] To achieve the above objectives, the technical solutions disclosed in the present invention are as follows:

[0005] According to one aspect of the present disclosure, a method for constructing a three-dimensional geotechnical engineering investigation information model based on BIM is provided, wherein the steps of the construction method include:

[0006] Acquire multi-dimensional geotechnical survey data. Collect geotechnical survey data from multiple sources in real time, including geological borehole location coordinates, soil sample properties, groundwater level dynamic values, and geological structure maps. During the collection process, perform data synchronization operations to ensure time stamp consistency.

[0007] Data preprocessing and standardization: cleaning the survey data, removing outliers, and converting the data format based on preset geotechnical engineering geology classification standards to generate a unified structured data set;

[0008] Initialize the multi-dimensional space framework based on the BIM model, start the BIM platform interface, input the unified structured data set, and construct a three-dimensional space reference coordinate system. The three-dimensional space reference coordinate system locates its origin by dynamically integrating geotechnical layer information and spatial geographic coordinates;

[0009] Parameter-driven geotechnical information model reconstruction: applying a parameter-driven mechanism in the three-dimensional spatial reference coordinate system, inputting dynamic weight vectors of geotechnical properties, and automatically generating a three-dimensional geotechnical engineering investigation information model in combination with geophysical constraints. The investigation information model includes geological body distribution, porosity parameters, and shear strength indicators.

[0010] Model output and application optimization: output the survey information model, use the rendering engine of the BIM platform for visualization, start a self-optimization feedback loop based on the engineering geological risk level, and apply the survey information model to the design decision support system.

[0011] Furthermore, the step of generating the unified structured data set includes:

[0012] setting an outlier detection threshold based on geotechnical engineering specifications, performing a filtering operation on the survey data to remove outliers and fill in missing data points;

[0013] Converting the filtered data into a BIM-compatible format, including encoding soil sample properties into IFC entity objects;

[0014] Logically assign data labels based on geological hierarchies to generate a geotechnical attribute matrix;

[0015] Adding timestamps and geological context descriptors to the geotechnical attribute matrix, constructing the unified structured data set, and performing quality assessment operations to calculate the data confidence score. If the score is lower than a preset threshold, reacquiring the survey data.

[0016] Furthermore, the model construction method also includes a dynamic field reconstruction of rock and soil heterogeneity to generate a high-precision rock and soil property field, the steps of which include:

[0017] Define the three-dimensional space position vector x and the target attribute function φ(x);

[0018] Design of multidimensional kernel function based on geomechanics principles ,in is the reference point vector;

[0019] Dynamic field integral operation, the expression is:

[0020] ,in, is the density distribution function, Γ(x) is the normalization factor, κ(x) is the heat conduction simulation coefficient, τ(x) is the stress field function, is the thermal-mechanical coupling coefficient;

[0021] The gradient descent method is used to solve the expression to minimize the error, and the reconstructed geotechnical property field is output to maintain the dynamic response capability of the geotechnical heterogeneous body model.

[0022] Furthermore, the step of constructing the three-dimensional space reference coordinate system includes:

[0023] dynamically calculating the position of the geological reference origin based on the unified structured data set;

[0024] Divide the multi-layer geotechnical space blocks based on the geological reference origin;

[0025] Dynamic data fusion factors are applied to adjust block weights to generate an adaptive three-dimensional spatial reference coordinate system, which is then interactively bound to the BIM platform interface to activate a real-time data update module.

[0026] Furthermore, the step of reconstructing the parameter-driven geotechnical information model includes:

[0027] extracting a geotechnical property dynamic weight vector from the unified structured data set;

[0028] Input geophysical constraint set, including rock and soil shear limit and porosity threshold;

[0029] Iteratively calculate 3D point cloud data to generate an initial model;

[0030] The three-dimensional point cloud data is compared with the measured data. If the error exceeds the tolerance, the dynamic weight vector of the geotechnical attribute is modified until the model accuracy meets the standard.

[0031] Furthermore, before outputting the survey information model, a model error correction and feedback mechanism is performed, the steps including:

[0032] Collecting the deviation value between the output point of the survey information model and the actual measured point of the geological drilling;

[0033] Calculate credibility intervals based on the Bayesian algorithm;

[0034] Feedback adjustment is initiated based on the credibility interval to optimize the parameter driving mechanism.

[0035] Furthermore, the step of obtaining the survey data includes:

[0036] The three-dimensional coordinates of multiple geological borehole locations, including longitude, latitude, and depth coordinates, are collected by a sensor array, and rock and soil samples of each borehole are continuously measured to generate an initial data stream;

[0037] Using a time synchronization protocol to align the initial data stream with a geographic information system layer in real time, and calculating a time delay compensation value to ensure data timing consistency;

[0038] A cross-validation operation is performed to compare the initial data stream with a preset geotechnical engineering threshold. If any data point exceeds the threshold range, a data re-collection module is triggered to perform redundant collection to output an enhanced multi-dimensional geotechnical engineering investigation dataset.

[0039] Furthermore, the step of outputting the survey information model includes:

[0040] Apply BIM rendering engines to perform 3D visualization and convert the model into interactive views;

[0041] Based on the practicality of the engineering geological risk level assessment model, when the risk level is high, a self-optimization cycle is triggered to update the model. At the same time, an output module is added to export the model to a cloud database format.

[0042] Furthermore, the parameter-driven geotechnical information model reconstruction step further includes:

[0043] Execute geotechnical stability analysis algorithms to calculate potential sliding surfaces based on 3D models;

[0044] Adjust analysis parameters through dynamic factors and generate stability reports.

[0045] According to another aspect of the present disclosure, a BIM-based 3D geotechnical engineering investigation information model construction system is provided, for implementing the BIM-based 3D geotechnical engineering investigation information model construction method described above, the construction system comprising:

[0046] Data acquisition module, used to collect multi-dimensional geotechnical engineering survey data in real time, including geological drill hole coordinates, soil property samples and groundwater level values;

[0047] a preprocessing module, connected to the data acquisition module, for cleaning, standardizing and transforming the data to generate a structured data set;

[0048] A BIM engine module, integrating the output of the preprocessing module, for initializing a three-dimensional spatial reference coordinate system and dynamically fusing layer information;

[0049] A model reconstruction module, interacting with the BIM engine module, for generating a three-dimensional geotechnical engineering investigation information model through a parameter-driven mechanism;

[0050] an output optimization module, connected to the model reconstruction module, for performing visualization processing, applying self-optimization feedback, and exporting model data;

[0051] The model reconstruction module includes:

[0052] Data fusion submodule, used to input geotechnical attribute dynamic weight vector;

[0053] The geological constraints submodule is used to integrate physical constraints;

[0054] an algorithm processor submodule, interacting with the data fusion submodule and the geological constraint submodule to perform a dynamic field integration operation;

[0055] Visualization submodule, used to display the reconstructed model.

[0056] The beneficial effects of the present invention are:

[0057] The present invention fundamentally breaks the technical bottleneck of static physical process in traditional modeling through the synergistic effect of the geotechnical information generation mechanism of multi-physical field coupling and the dynamic response architecture of the BIM framework. Specifically, the present invention constructs a thermal-mechanical dynamic feedback closed loop under geomechanical constraints: in the parameter-driven reconstruction stage, the dynamic weight vector of geotechnical properties not only integrates conventional parameters such as porosity and shear strength in the three-dimensional spatial coordinate system, but also embeds the differential constraint association of groundwater flow field-temperature gradient-stress tensor, so that the time-varying degradation behavior of heterogeneous rock and soil bodies is accurately mapped to the adaptive adjustment of model parameters. This mechanism ensures that when the dynamic water level value is input into the survey data, the system automatically triggers the recalculation of the seepage path through the physical field driven kernel function, and simultaneously corrects the spatial probability model of the geological body distribution, thereby converting the geological energy transfer law contained in the drilling data into a quantifiable engineering response.

[0058] During the output phase, a self-optimizing feedback loop based on geological risk and the dynamic coupling of the decision support system form a two-way verification chain: Model visualization and real-time analysis of the gradient changes in geotechnical shear strength indicators under the influence of thermal-mechanical coupling. When engineering disturbances cause rock formation stresses to exceed a critical threshold, the self-optimization mechanism immediately feeds back to the parameter-driven layer to reconstruct the weight vector, ensuring that the output model always maintains thermodynamic homeomorphism with the geological prototype. This architecture represents a breakthrough in compressing the dynamic response of the entire geotechnical engineering lifecycle into a unified computational process, eliminating the evaluation distortion caused by the separation of data-driven and physical mechanisms in traditional BIM modeling, and providing a digital twin with strict thermoelastic constraints for engineering decision-making in complex geological environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 Flowchart of a method for constructing a three-dimensional geotechnical engineering investigation information model based on BIM in one embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of the thermomechanical coupling field integration principle in one embodiment of the present invention;

[0061] Figure 3 Schematic diagram of multidimensional geological kernel function distribution in one embodiment of the present invention;

[0062] Figure 4Schematic diagram of the rock formation occurrence gradient field in one embodiment of the present invention;

[0063] Figure 5 A schematic diagram of constructing a three-dimensional spatial reference coordinate system according to an embodiment of the present invention;

[0064] Figure 6 Schematic diagram of dynamic weight optimization of geotechnical properties in one embodiment of the present invention;

[0065] Figure 7 A schematic diagram of Bayesian credibility interval analysis and model correction in one embodiment of the present invention;

[0066] Figure 8 Schematic diagram of a geotechnical static interpolation model in one embodiment of the present invention;

[0067] Figure 9 Schematic diagram of a geothermal-mechanical coupling dynamic model in one embodiment of the present invention;

[0068] Figure 10 Schematic diagram of a three-dimensional sliding surface for geotechnical stability analysis in one embodiment of the present invention. DETAILED DESCRIPTION

[0069] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0070] The present invention provides the following preferred embodiments:

[0071] Example 1: In order to solve the problem of insufficient accuracy of three-dimensional model construction caused by the multi-source heterogeneity of geotechnical engineering survey data, this example discloses a method for constructing a geotechnical engineering survey information model based on BIM, which realizes accurate three-dimensional expression of geological attributes through systematic data fusion and spatial parameterization driving mechanism. Figure 1 As shown, the specific process is as follows:

[0072] S100: Acquire multi-dimensional geotechnical survey data. This data is collected from multiple sources in real time, including geological borehole location coordinates, soil sample properties, groundwater level dynamic values, and geological structure maps. During the collection process, data synchronization is performed to ensure timestamp consistency.

[0073] S200: Data preprocessing and standardization: cleaning the survey data, removing outliers, and converting the data format based on the preset geotechnical engineering geology classification standards to generate a unified structured data set.

[0074] S300: Initialize the multi-dimensional space framework based on the BIM model, start the BIM platform interface, input the unified structured data set, and build a three-dimensional spatial reference coordinate system. The three-dimensional spatial reference coordinate system locates the origin by dynamically integrating geotechnical layer information and spatial geographic coordinates.

[0075] S400: Parameter-driven geotechnical information model reconstruction applies a parameter-driven mechanism in a three-dimensional spatial reference coordinate system, inputs dynamic weight vectors of geotechnical properties, and combines geophysical constraints to automatically generate a three-dimensional geotechnical engineering survey information model. The survey information model includes geological body distribution, porosity parameters, and shear strength indicators.

[0076] S500: Model output and application optimization, output the survey information model, use the BIM platform's rendering engine for visualization, and start a self-optimization feedback loop based on the engineering geological risk level to apply the survey information model to the design decision support system.

[0077] Furthermore, during the data acquisition phase, raw survey data is collected simultaneously from independent data sources, including geological radar detection systems, borehole sampling equipment, and groundwater monitoring sensors. It is important to understand that this multi-source data includes the three-dimensional spatial coordinates (X, Y, Z) of geological boreholes, physical and mechanical parameters of soil samples (such as moisture content and particle size distribution), dynamic groundwater level monitoring time series, and geological structural vector graphics. During the acquisition process, millisecond-level clock synchronization is achieved through the NTP protocol, ensuring that all survey data timestamps are standardized to a standard format, eliminating spatial registration errors caused by time sequence misalignment.

[0078] Furthermore, multidimensional standardization was performed during the data preprocessing phase. The collected raw data was first filtered for outliers, using the Grubbs criterion to remove outliers that deviated more than three standard deviations from the sample mean. It is important to understand that outliers primarily occur in records of sudden changes in groundwater levels and sudden changes in borehole depth. Subsequently, the unstructured data was converted into a unified structured dataset based on the Class III geological classification standard. This dataset was stored in a relational database table, with fields including spatial coordinates (EPSG:4979 coordinate system), geological time (classified by era, epoch, and epoch), and 17 geophysical property fields.

[0079] Furthermore, during the BIM spatial framework construction phase, the Autodesk Revit API is called to create a 3D reference coordinate system. It is understandable that the origin of the coordinate system is determined by dynamically integrating the geological layer control points and the geographic coordinates: the coordinates of the BH-01 borehole in the field are used as the temporary origin (x0, y0, z0), the elevation of the bedrock top is inverted through Kriging interpolation, and finally the origin is vertically projected to the top of the bedrock to form a permanent origin (x0, y0, z0). bedrockThe axial definition of the coordinate system follows the right-hand rule, where the X-axis is parallel to the main structural line, with an azimuth of 112°, and the Z-axis is vertically downward as the positive direction.

[0080] Furthermore, when implementing parameter-driven model reconstruction, the geotechnical attribute dynamic weight vector W = [w p ,w s ,w d ] T It is important to understand that w p The spatial weight coefficient (0.42-0.78) characterizing the porosity property, w s Control shear strength index variation coefficient (0.15-0.33), w d is the density distribution gradient factor. Physical constraints are applied simultaneously during the reconstruction process, such as the Mohr-Coulomb failure criterion defining the shear strength envelope and Darcy's law constraining the porosity-permeability conversion relationship. A weighted overlay algorithm is used to generate a three-dimensional model that includes the spatial distribution of the geological volume, porosity isosurfaces, and the shear strength gradient field.

[0081] The benefit of this embodiment lies in the integrated expression of all elements of survey information. After generating a visual model through rendering using the BIM platform's NURBS surface engine, the system triggers a self-optimization feedback loop based on a preset engineering geological risk level, such as landslide risk level II. This loop dynamically adjusts the borehole data weights within the decision support system, stabilizing the model porosity prediction error within a ±5% threshold. This closed-loop optimization mechanism ensures the timeliness and reliability of the geological parameters on which design decisions rely. Through the implementation of this embodiment, the shortcomings of traditional modeling, such as the inconsistent spatiotemporal benchmarks of multi-source data and the discretization of geological attribute expression, are overcome, providing a high-precision three-dimensional information foundation for engineering design and risk prevention and control under complex geological conditions.

[0082] Example 2: To address the problem of insufficient model construction accuracy due to outlier interference and format heterogeneity in geotechnical engineering survey data, this example refines the process of generating a unified structured data set, and ensures deep compatibility between the survey data and the BIM platform through standardized data cleaning, format conversion, logical label allocation, and quality control.

[0083] Specifically, outlier detection rules are set based on geotechnical engineering industry specifications, and filtering operations are performed on the original survey data. The system pre-configures multi-level threshold intervals, and uses a dynamic sliding window algorithm to identify outliers for geotechnical parameters in the drilling data, such as moisture content, compressive strength, etc. - when a data point continuously deviates from the mean of the same layer by more than a preset standard deviation multiple, it is marked as an outlier and removed. For missing data points, the inverse distance weighted interpolation method in geological statistics is used to fill them. This method reconstructs missing values ​​through the weighted average of the attribute values ​​of surrounding valid data points based on the principle of spatial proximity. The weight is inversely proportional to the distance to ensure that the filled data meets the continuity characteristics of the geological layer. It should be understood that this step effectively improves the integrity and reliability of the data through regularized anomaly detection and reasonable interpolation strategies, laying the foundation for subsequent processing.

[0084] Furthermore, the filtered data is converted into a format compatible with the BIM platform. The core lies in encoding the soil sample properties into IFC (Industry Foundation Classes) standard entity objects. The system establishes a data mapping rule library to correspond physical properties such as geotechnical type, density, and permeability coefficient to specific entities or attribute sets in the IFC model: for example, the soil layer division results are mapped to the IFC entity, and its mechanical parameters are mounted to the IfcPropertySet extended attributes to form a structured data unit that can be recognized by mainstream BIM software. At the same time, for spatial information such as borehole coordinates and stratum depth, a geometric association with the IFC spatial structure tree is established to ensure the accurate positioning of the data in the three-dimensional model. Through this standardized conversion, the format barrier between traditional survey data and the BIM platform is resolved, allowing geotechnical engineering parameters to be directly integrated into the three-dimensional information model.

[0085] After completing data cleaning and format conversion, the data is labeled according to the geological hierarchy logic to generate a geotechnical attribute matrix. Based on geological features such as stratigraphic sedimentary sequence and lithologic boundaries, the system divides the survey data into different hierarchical units, such as surface fill, silty clay, sandstone, etc. Each unit is assigned a unique logical label, which contains information such as stratigraphic number, lithologic code, and burial depth interval. On this basis, the geotechnical properties of each layer, such as porosity, cohesion, and compression modulus, are organized into a matrix form according to the spatial coordinate order. The rows of the matrix correspond to different geological layers, and the columns correspond to attribute categories and spatial position parameters, such as X, Y coordinates, and elevation. This structured matrix expression not only clearly reflects the vertical stratification characteristics of geotechnical properties, but also facilitates the subsequent parametric association with the BIM model, providing a direct data interface for the attribute assignment of the three-dimensional model.

[0086] Finally, add timestamps and geological context descriptors to the geotechnical attribute matrix, build a unified structured data set, and perform quality assessment operations simultaneously. The timestamp records the specific time of data collection and is used to trace the timeliness of the data; the geological context descriptor contains qualitative information such as stratum occurrence (such as rock layer dip, inclination), structural characteristics (such as fault distribution, fold morphology), etc., which are recorded by combining natural language with structured coding. For example, the fault position is marked as "F1 fault, strike NE30°, dip 65°" and associated with the matrix unit of the corresponding layer. Figure 7 As shown in Figure 1, the quality assessment module calculates a data confidence score using a preset algorithm. This score takes into account factors such as data integrity, the proportion of valid data, standardization, and spatiotemporal consistency. If the confidence score falls below a preset threshold, the system automatically triggers the data re-acquisition process, prompting surveyors to review or retest abnormal data points to ensure that the data entering the model construction phase meets the accuracy requirements.

[0087] It is understandable that the timestamp mechanism provides a time dimension traceability basis for data version management, while the geological context descriptor supplements the geological background information that cannot be expressed by pure numerical data. The two together with the attribute matrix constitute a multi-dimensional structured data system. The quality assessment link realizes automatic control of data quality through quantitative indicators to avoid model errors caused by unreliable data. Through the above steps, this embodiment converts the original survey data into a standardized data set with spatial positioning, attribute description, time attribute and quality identification, providing a standardized and reliable data foundation for the subsequent construction of a three-dimensional model based on BIM, ensuring that the model can accurately reflect the actual geological conditions of geotechnical engineering and enhance the engineering application value of the information model.

[0088] Example 3: This example provides a method for constructing a three-dimensional geotechnical engineering survey information model based on BIM. Figures 2 to 4 、 Figure 5 and Figure 8 and Figure 9 As shown in the figure, the steps of reconstructing the dynamic field of rock and soil heterogeneity are specifically described to achieve high-precision rock and soil property field generation.

[0089] Specifically, if Figure 5 As shown in Figure 1, a three-dimensional spatial reference coordinate system O-XYZ is established with the engineering origin of the geological survey area as the coordinate origin, where the X-axis points due east (longitude), the Y-axis points due north (latitude), and the Z-axis points in the positive elevation direction (opposite to depth). Define the spatial position vector x = (x, y, z), and the target property function φ(x) represents the target physical properties of the geotechnical mass (such as elastic modulus, permeability, etc.).

[0090] Combine Figure 5The spatial distribution characteristics of the multi-layer rock and soil layers and fault structures are used to discretize the survey area Ω and generate a uniform grid point set {x i}, where i = 1, 2, …, N, where N is the total number of grids, ensuring that the coordinate system covers all regional geological interfaces (such as rock and soil layer boundaries, fault zones, etc.).

[0091] Furthermore, based on the principles of geomechanics, we construct Figures 2 to 4 The multidimensional kernel function shown , used to quantify the spatial point x and the reference point The attribute correlation between Figures 2 to 4 Kernel function visualization results):

[0092] Geological anisotropy kernel: The Gaussian kernel is used to characterize the attenuation relationship between lithologic differences and spatial distance. The expression is:

[0093] , where α is the anisotropic strength coefficient and σ1 is the spatial correlation scale, corresponding to Figures 2 to 4 The solid line "geological anisotropy core ", and its curve shape reflects the characteristics of strong correlation in the near field and weak correlation in the far field.

[0094] Stress conduction kernel: The long-range attenuation kernel is used to describe the spatial conduction effect of the tectonic stress field. The expression is:

[0095] , where β is the stress transmission coefficient, σ2>σ1, corresponding to Figures 2 to 4 The middle dotted line is the “stress conduction core”, and its gentle decay characteristics reflect the long-distance influence of the stress field.

[0096] The final kernel function is the weighted sum of the two:

[0097] ;

[0098] like Figures 2 to 4 As shown in the schematic diagram of the stress-heat conduction coupling field, the property field reconstruction formula including the density field integral term and the thermal-mechanical coupling term is constructed:

[0099] ,in, is the density distribution function, Γ(x) is the normalization factor, κ(x) is the heat conduction simulation coefficient, τ(x) is the stress field function, is the thermal-mechanical coupling coefficient.

[0100] Density field integral term: through kernel function Density distribution function in region Ω Perform weighted integration, Γ(x) is the normalization factor to ensure the conservation of attribute field energy, corresponding to Figures 2 to 4 Visualization of kernel function distribution and integration principle.

[0101] Stress-heat conduction coupling term: The heat conduction effect κ(x) is dynamically related to the tectonic stress field τ(x) through the heat-mechanical coupling coefficient λ. Among them, κ(x) reflects the heat conduction capacity of the rock mass, such as the κ difference between aquifers and dense rock layers. τ(x) is obtained by inversion of ground stress monitoring data. The gradient operator Characterize the coupling effect of heat flow and stress gradient, and solve the problem that traditional interpolation models ignore the interaction of physical fields, such as Figure 8 and Figure 9 Comparison between the "static interpolation model" and the "thermal-mechanical coupled dynamic model". The dynamic model can capture the anomalies in the properties of fault zones and lenses.

[0102] Furthermore, if Figure 6 Schematic diagram of dynamic weight optimization of geotechnical properties, defining the reconstruction error function:

[0103] , where φ meas (x i ) is the measured attribute value, λ reg is the regularization coefficient to suppress overfitting. The kernel function parameters {α, β, σ1, σ2} and the coupling coefficient λ are iteratively optimized by the gradient descent method. The update formula is:

[0104] , where θ is the parameter to be optimized and η is the learning rate. Physical constraints are introduced during the iteration process, such as Figure 6 The constraint equations in , ensure that the reconstructed field satisfies the geomechanical equilibrium condition. When the error converges, such as Figure 6 When the convergence point after 2.5 iterations is reached and the weight curve is stable, the final reconstructed geotechnical property field φ(x) is output. This field can dynamically respond to changes in thermal-mechanical coupling conditions, such as groundwater flow velocity and ground stress fluctuations, and maintain the model's high-precision characterization of the spatial distribution of heterogeneous bodies. Figure 8 and Figure 9 The medium dynamic model significantly suppresses the fault zone attribute values ​​and locally enhances the lens.

[0105] By using the method of this embodiment, compared with the traditional static interpolation model, such as Figure 8 and Figure 9 Left: Dynamic field reconstruction model, e.g. Figure 8 and Figure 9 The right figure can accurately capture the complex characteristics of geotechnical heterogeneity, such as nonlinear boundaries and local anomalies, and reflect the influence of physical field interaction on property distribution through thermal-mechanical coupling terms, which is helpful for subsequent three-dimensional sliding surface stability analysis, such as Figure 10 Critical sliding surface positioning provides more reliable basic data, significantly improving the engineering applicability and dynamic analysis capabilities of geotechnical engineering investigation models.

[0106] Example 4: In order to solve the problem that the existing three-dimensional spatial reference coordinate system is difficult to dynamically adapt to multi-source heterogeneous data in a complex geological environment, this example further refines the specific implementation method of constructing a three-dimensional spatial reference coordinate system. By dynamically calculating the geological reference origin, adaptively dividing the geotechnical space blocks and integrating the weights of multi-source data, an intelligent coordinate system that deeply interacts with the BIM platform is formed.

[0107] First, for the dynamic calculation of the geological reference origin, the implementation process requires the integration of multiple sources of data, such as borehole exploration data, seismic reflection interfaces, and surface topographic control points. Specifically, a coordinate optimization model based on the entropy weight method is used to determine the origin position. Its core calculation formula is:

[0108] , where O represents the three-dimensional coordinates of the geological reference origin, (x k ,y k ,z k ) is the spatial coordinate of the kth basic data point, m is the total number of data points involved in the calculation, w k is the data point weight coefficient. This weight coefficient is calculated by the entropy weight method and comprehensively reflects the monitoring accuracy, geological layer representativeness and spatial distribution density of the data point. The specific expression is:

[0109] , where e k is the information entropy of the kth data point, p ki is the normalized value of the kth data point in the i-th evaluation indicator, such as borehole depth error or seismic velocity signal-to-noise ratio, where n is the evaluation indicator dimension. This model allows the origin position to dynamically respond to the spatiotemporal heterogeneity of geological data, ensuring that the coordinate system reference point is located in the relatively stable core area of ​​the regional geological structure.

[0110] Furthermore, when dividing multi-layer geotechnical spatial blocks based on the geological reference origin, it is necessary to base the division on the spatial distribution characteristics of the stratigraphic sedimentary sequence and the tectonic interface. First, a 3D geological modeling algorithm, such as the horizon tracing technique based on inverse distance weighting, is used to extract the top and bottom interfaces of each geotechnical layer to form a layered spatial segmentation surface. For the lth geotechnical layer, its spatial range can be expressed as:

[0111] ;

[0112] Among them, z l top (x,y) and z l bottom (x, y) are the elevation functions of the top and bottom surfaces of the lth layer, respectively, generated by the geological interface interpolation algorithm; D lThe domain is defined by the projection of the horizon onto the xy plane, defined by geological structural features such as the stratigraphic pinchout line and fault boundaries. During the block demarcation process, a horizon attribute database is established, integrating multi-dimensional attributes such as geotechnical parameters (such as elastic modulus and Poisson's ratio) and permeability coefficient to form an attributed spatial grid structure.

[0113] When applying the dynamic data fusion factor to adjust the block weight, a three-dimensional weight adjustment model is constructed, which includes the time decay factor, spatial correlation coefficient and data credibility index. Specifically, the comprehensive weight w of the lth block at time t is l (t) is defined as:

[0114] ,in, is a time decay function, reflecting the dynamic correction of the block weight by the latest survey data. is the time attenuation coefficient, t0 is the reference time point; c l is the initial credibility coefficient of the lth layer, which is comprehensively evaluated by parameters such as survey data density and test method accuracy; d l is the Euclidean distance from the block center to the geological reference origin, σ is the spatial influence radius parameter, determined through geostatistical analysis, and L is the total number of geotechnical spatial blocks. This weighting model adaptively adjusts the contribution of each layer in the coordinate system construction, dynamically focusing the coordinate system on areas with frequent data updates or complex geological conditions.

[0115] Furthermore, the generated adaptive three-dimensional spatial reference coordinate system is interactively bound to the BIM platform interface, and real-time mapping of the coordinate system and model components is achieved through a customized data interface. Specifically, it includes: establishing a coordinate system parameter database to store core data such as origin coordinates, block division parameters and weight factors; developing a two-way data synchronization module to ensure that the geometric model and attribute information in the BIM platform are consistent with the dynamic update of the coordinate system; designing a visual interactive interface to support users to view the coordinate system grid division, weight distribution and data update status in real time in the BIM environment. When new survey data such as real-time monitored groundwater level and stress data are connected, the system automatically triggers the dynamic data fusion process, recalculates the block parameters through the weight model, and synchronously updates the spatial positioning and attribute association in the BIM model to achieve spatiotemporal consistency management of survey information.

[0116] It is important to understand that during the above implementation process, the dynamic calculation of the geological reference origin provides an adaptive benchmark positioning for the coordinate system, the multi-layer block division realizes the structured expression of the geological space, and the dynamic data fusion factor gives the coordinate system the ability to dynamically respond to multi-source data. The three form an organic whole through data flow and algorithm coupling, ensuring that the three-dimensional spatial reference coordinate system can reflect the spatiotemporal variation characteristics of geological conditions in real time. It is understandable that the deep binding of this coordinate system with the BIM platform not only provides a precise spatial positioning framework for geotechnical engineering survey information, but also builds an infrastructure for multi-source data integration and dynamic updating, providing standardized data interfaces and spatial benchmarks for subsequent functional modules such as numerical simulation and stability analysis.

[0117] The benefit of this embodiment lies in the fact that by constructing a 3D spatial reference coordinate system that incorporates dynamic datum positioning, hierarchical structure division, and data fusion weighting, it effectively solves the spatial integration challenge of multi-source survey data in complex geological environments, forming an intelligent coordinate system that deeply collaborates with the BIM platform. This system can dynamically adjust spatial division and weighting based on real-time data, ensuring that the survey information model remains synchronized with actual geological conditions, providing key technical support for the intelligent and refined development of 3D geotechnical engineering surveys.

[0118] Example 5: To address the issue of accurately reconstructing parameter-driven geotechnical information models under complex geological conditions, this example further refines the collaborative mechanism of dynamic weight adjustment and constraint verification during model construction. Specifically, this parameter-driven geotechnical information model reconstruction method, based on a unified structured dataset, achieves precise characterization of the spatial distribution characteristics of three-dimensional geological bodies through the dual control of adaptive adjustment of dynamic weight vectors and geophysical constraints.

[0119] Specifically, a dynamic weight vector of geotechnical properties, including core attributes such as shear strength, porosity, and density, is extracted from a unified structured data set. This vector quantifies the differences in the degree of influence of each geotechnical attribute in model construction, providing a differentiated weight distribution benchmark for subsequent iterative calculations. It should be understood that the dynamic weight vector is not fixed, but is based on the spatial variation characteristics of the geological body and the engineering analysis objectives. The contribution of each attribute to the model accuracy is dynamically determined through a data mining algorithm, thereby forming a weight distribution scheme that reflects the current geological conditions. For example, in areas of high stress concentration, the weight value of the shear strength attribute will be automatically increased based on the inversion results of historical data to highlight the key role of this attribute in model stability analysis.

[0120] Furthermore, a set of geophysical constraints including physical boundary conditions such as the shear limit of the rock and soil layer and the porosity threshold is input. These constraints define scientific boundaries for the value range of each attribute parameter in the model construction process by integrating geotechnical theory and engineering specifications. For example, the shear limit of the rock and soil layer is determined based on the Mohr-Coulomb strength theory, and the porosity threshold is set based on the sedimentary environment of the stratum and the rock genesis type, ensuring that the model parameters conform to the objective physical properties of the geological materials and avoiding the generation of unreasonable results that violate the basic mechanical principles. During the iterative calculation process, the above constraints serve as mandatory verification standards to monitor the evolution path of each attribute parameter in real time. Once a parameter exceeds the limit, the weight adjustment mechanism is triggered.

[0121] In the iterative calculation phase of three-dimensional point cloud data, the extracted dynamic weight vectors and geophysical constraints are deeply integrated into the model construction algorithm. Specifically, by constructing an objective function containing weight parameters, the spatial distribution characteristics of geotechnical properties are converted into quantifiable mathematical expressions, and the spatial interpolation weight ratio of each attribute is adjusted successively according to the current weight distribution in each iteration. At the same time, each step of the calculation needs to verify whether the preset physical constraints are met. For example, when the calculated value of the porosity parameter exceeds the threshold, the algorithm automatically reduces the weight contribution of the density attribute and enhances the coupling correlation between the porosity and shear strength attributes, thereby guiding the model parameters to converge to a reasonable range. Through the synergistic effect of this weight adjustment and constraint verification, an initial model that preliminarily reflects the spatial distribution characteristics of the geological body is gradually generated.

[0122] After the initial model is generated, the measured data verification phase begins. The three-dimensional point cloud data output by the model is compared for accuracy with the measured geological data obtained through drilling, geophysical exploration, and other means. Quantitative indicators such as spatial interpolation error and attribute parameter mean square error are used to evaluate the model fitting effect. If the comparison result shows that the error exceeds the preset threshold, the system automatically returns to the weight extraction phase and adaptively adjusts the dynamic weight vector of geotechnical properties. The adjustment strategy is dynamically determined based on the error distribution characteristics. For example, in areas where the shear strength parameter error is significant, the attribute weight is strengthened and the influence of secondary attributes is weakened, guiding the model to focus on the fitting optimization of key geological features. This closed-loop feedback mechanism forms a cyclic optimization process of "iterative calculation-error comparison-weight adjustment" until the model accuracy meets the preset standards for engineering applications.

[0123] It should be noted that, in the above reconstruction process, if Figure 5 As shown in , the three-dimensional spatial reference coordinate system provides a unified geographic spatial benchmark for model construction, ensuring the spatial registration accuracy of data from different sources; Figures 2 to 4As shown in , the thermodynamic coupling field integral principle provides a physical mechanism support for the generation of dynamic weight vectors, so that the weight distribution is not only based on the statistical characteristics of the data, but also in line with the thermodynamic evolution law of the geological body. In addition, in areas with complex geological structures such as faults and lenses, the dynamic weight vector can be locally adaptively adjusted according to the spatial distribution characteristics of geological heterogeneity, further improving the model's ability to characterize heterogeneous geological bodies, such as Figure 8 and Figure 9 shown.

[0124] The benefit of this embodiment lies in the organic combination of dynamic weight adjustment of geotechnical properties with geophysical constraints, which constructs an adaptive and physically rational model reconstruction framework. This framework not only utilizes data-driven methods to explore implicit geological laws, but also ensures that the model conforms to basic mechanical principles through physical constraints. This effectively addresses the problems of insufficient fitting accuracy or lack of parameter rationality of traditional models under complex geological conditions, providing reliable basic data support for subsequent geotechnical stability analysis and engineering design. Through this systematic reconstruction method, accurate mapping of actual geological conditions is achieved, laying a solid foundation for the engineering application of three-dimensional geotechnical engineering survey information models.

[0125] Example 6: To address the issue of geological data uncertainty affecting the accuracy of survey information model output, this example further refines the specific implementation of the model error correction and feedback mechanism. By constructing a Bayesian credibility analysis framework, this mechanism achieves targeted optimization of the parameter-driven mechanism. Specifically, this mechanism quantitatively analyzes the deviation between measured data and model outputs to establish parameter adjustment rules under probabilistic constraints, ensuring dynamic adaptation of the model parameter space to actual geological conditions.

[0126] First, in the deviation value collection stage, the output points of the survey information model and the actual measured points of the geological drilling are unified to Figure 5 In the three-dimensional spatial reference coordinate system shown, a one-to-one correspondence between spatial data pairs is formed. It is important to understand that the measured borehole points contain key geotechnical properties such as strength parameters, porosity, and density, while the model output points are derived from continuous field data generated by the parameter-driven mechanism. During this process, the system automatically filters out abnormal deviation points caused by measurement errors and uses robust statistical methods to calculate the sample mean and standard deviation of each attribute parameter to form the initial deviation dataset.

[0127] In the Bayesian credibility interval calculation stage, a probability model including prior distribution, likelihood function and posterior distribution is constructed. The prior distribution is based on the regional geological survey experience data and defines the reasonable fluctuation range of geotechnical parameters, such as reference Figure 6In the dynamic weight optimization process of geotechnical properties, the prior distribution of shear strength weight is set to normal distribution, whose mean and variance correspond to the statistical characteristics of historical engineering data. The likelihood function describes the probability of the observed data under given model parameters. Figure 7 The error distribution histogram analysis shown in Figure 1 uses kernel density estimation to fit the probability density function of the measured deviation to form a probabilistic description of the model prediction error. Furthermore, the Markov Chain Monte Carlo (MCMC) algorithm is used to sample the posterior distribution and calculate the 95% confidence interval, which represents the reasonable range of model parameters under the existing data and prior knowledge. When the measured value exceeds the confidence interval of the model output, the feedback adjustment mechanism is triggered, such as Figure 7 The system automatically identifies the abnormal areas marked in the image as key areas that require parameter optimization.

[0128] In the process of optimizing the parameter-driven mechanism, the key parameters in the model are adjusted hierarchically according to the analysis results of the credibility interval. Figure 6 The dynamic weight optimization module shown in the figure adjusts the weight coefficients of geotechnical properties, such as increasing the weight of borehole measured data in parameter interpolation to reduce the uncertainty of model prediction values. Specifically, in the 3D geological modeling process, for the geotechnical density parameters in abnormal areas, the bandwidth of the spatial kernel function in the Kriging interpolation is reduced by 20%, thereby enhancing the constraint effect of local measured data on the model output. At the same time, combined with Figures 2 to 4 The thermomechanical coupling field integration principle shown in the figure starts an adaptive correction algorithm in the area outside the credibility interval for parameters involving thermomechanical coupling effects, such as thermal conductivity and stress conduction coefficient. The shape parameters of the kernel function are adjusted through iterative optimization, so that the model output value gradually converges to the statistical distribution range of the measured data.

[0129] It should be noted that the feedback adjustment process follows the principle of gradual optimization to avoid model instability caused by drastic parameter changes. The system sets a dual-threshold control mechanism: when the deviation value exceeds the upper limit of the credibility interval but is less than 1.5 times the interval width, a slight adjustment is initiated, and only the parameter weights of the local grid are fine-tuned; when the deviation value exceeds 1.5 times the interval width, a deep optimization is triggered to rebuild the parameter-driven network in the area and introduce Figure 8 and Figure 9 The heterogeneous body model comparison results shown above select the dynamic model parameter combination that best fits the measured data. For example, after identifying that the porosity parameters near a fault zone consistently deviate from the confidence interval, the system automatically calls a modified kernel function that incorporates the fault's influence, replacing the default isotropic kernel function. By adjusting the directional parameters of the geological gradient field, the model more accurately reflects the fault's control over porosity distribution.

[0130] During implementation, the mechanism Figure 10 The three-dimensional stability analysis module, shown in Figure 1, forms a closed data loop, feeding error-corrected model parameters into the stability calculation module in real time to verify the changing trends of the sliding surface position and safety factor. This ensures that parameter adjustments not only reduce data deviations but also conform to the fundamental laws of geotechnical mechanics. For example, when the cohesion parameter in a certain area increases after correction, the corresponding three-dimensional sliding surface depth should also increase. The safety factor calculation results must meet the inherent consistency of the Mohr-Coulomb strength criterion. If abnormal fluctuations occur, the prior assumptions used in the confidence interval calculation process are reversely checked to ensure they are reasonable, forming a multi-verification mechanism.

[0131] The benefit of this embodiment lies in the deep coupling of Bayesian credibility analysis with a parameter-driven mechanism, establishing a complete closed loop from data deviation identification to model parameter optimization, achieving quantitative control of the uncertainty of the survey information model. This mechanism not only utilizes prior knowledge of historical geological data to constrain model output, but also adaptively adjusts parameter interpolation rules under complex geological conditions through dynamic feedback from measured data, ensuring the model's prediction accuracy in areas with heterogeneous rock and soil distribution, and providing reliable basic data support for subsequent engineering stability analysis and design.

[0132] Example 7: To address issues such as spatial coordinate deviation, time series asynchrony, and insufficient data reliability during survey data acquisition, this example further refines the specific implementation method for survey data acquisition, achieving the construction of an enhanced multidimensional geotechnical engineering survey dataset through the collaborative use of multiple technologies. First, a sensor array consisting of fiber optic sensors, resistance strain gauges, and a high-precision positioning module is deployed at the geological drilling site. Three-dimensional coordinate information, including longitude, latitude, and depth coordinates based on the geoid, is synchronously collected for each drilling point. A series of physical and mechanical parameters, such as density, moisture content, and elastic modulus, of the geotechnical samples are acquired through high-frequency continuous measurement, forming an initial data stream containing spatial location information and attribute characteristics. It should be understood that the positioning module uses real-time kinematic differential (RTK) technology for coordinate calculation, and depth coordinate measurement is combined with a borehole inclinometer for inclination correction to ensure that the three-dimensional coordinate accuracy meets the requirements of engineering survey specifications.

[0133] Furthermore, to achieve spatiotemporal alignment of multi-source data, a time synchronization system is constructed using the Network Time Protocol (NTP). This system calibrates the sensor array's acquisition clocks in real time with the spatiotemporal reference of the Geographic Information System (GIS). A two-way time transfer algorithm is used to calculate the time delay compensation value for each sensor node. This compensation value is dynamically updated based on the signal transmission distance, medium propagation speed, and device clock drift rate, eliminating the time misalignment of data streams caused by asynchronous sampling clocks and ensuring precise temporal alignment of the initial data stream with the GIS layer. It is understood that the GIS layer is pre-loaded with basic geographic information such as regional geological structure maps, stratigraphic contours, and the distribution of existing engineering facilities, providing a benchmark framework for subsequent data spatial analysis.

[0134] In the data quality control link, a cross-validation mechanism is established to perform real-time verification of the initial data stream. Specifically, the geotechnical parameter value corresponding to each data point is compared with the stratum characteristic database of the survey area. The database contains the geotechnical engineering threshold range preset according to the regional geological survey report, covering the reasonable range of key parameters such as shear strength, compression modulus, and permeability coefficient. If any data point is detected to be outside the threshold range, the system automatically triggers the data re-collection module, performs secondary collection through redundantly arranged sensors, and uses the Kalman filter algorithm to fuse the multiple collected data, filter out outliers, and output an enhanced data set after error correction. It should be noted that the data re-collection module has set up a three-level verification mechanism, which starts the on-site equipment fault diagnosis process for data points that exceed the threshold three times in a row to avoid continuous data distortion due to sensor hardware abnormalities.

[0135] The benefits of this embodiment lie in the multi-dimensional continuous acquisition of borehole data achieved through the three-dimensional deployment of sensor arrays. The deep integration of the time synchronization protocol and the GIS system ensures the consistency of the data's spatiotemporal benchmarks, while the cross-validation and redundant acquisition mechanisms based on threshold comparisons effectively improve the reliability of the survey data. The enhanced multi-dimensional geotechnical engineering survey dataset generated by this process not only fully preserves the spatial distribution characteristics and physical and mechanical properties of the geotechnical mass, but also ensures the engineering applicability of the data through error correction and quality control. This provides a high-precision data foundation for the subsequent construction of a BIM-based three-dimensional information model, enabling the model to more realistically reflect the geological conditions of the survey area and providing a reliable basis for geotechnical engineering design and stability analysis.

[0136] Example 8: In order to solve the problems of insufficient interactivity and dynamic adaptability in visualization during the output of the survey information model, this example further refines the specific implementation method for outputting the survey information model. First, the BIM rendering engine is used to perform visualization processing on the constructed three-dimensional geotechnical engineering survey information model. The rendering engine supports model rendering at multiple levels of precision and can dynamically adjust display details according to user operation requirements. For example, it automatically improves grid accuracy in areas with complex geological structures to ensure that the spatial features and attribute information of the three-dimensional model are intuitively presented. Through the engine's built-in interactive development interface, the model is converted into an interactive view, allowing users to obtain geological parameters of specific locations through operations such as coordinate picking, attribute query, and section cutting, forming an information display interface for human-computer collaboration.

[0137] Furthermore, in order to improve the model's dynamic response capability to engineering geological risks, a risk level assessment module is introduced to conduct real-time analysis of the survey data carried by the model. Based on a preset risk assessment indicator system, this module comprehensively calculates parameters such as rock and soil stability, permeability characteristics, and load adaptability to generate corresponding risk level results. When the assessment results determine that the risk level is high, the system automatically triggers the self-optimization cycle mechanism, and iteratively updates the model by tracing back the data collection source, verifying modeling parameters, and correcting the calculation logic to ensure that the model always reflects the latest risk status of the survey area. It should be understood that the self-optimization cycle process adopts an incremental update strategy, and only reconstructs local models with high risk correlation to avoid waste of computing resources caused by repeated processing.

[0138] For model output, a standardized data export module has been added. This module supports conversion to various cloud database formats, including but not limited to GeoJSON, CityGML, and the BIM-specific IFC format. During the export process, the system automatically performs lightweight processing on the model data, compressing the data volume while preserving key geometric features and attribute information, facilitating storage, sharing, and collaborative applications on the cloud platform. The exported cloud database format adheres to industry data exchange standards, ensuring compatibility with subsequent engineering design, construction management, and other systems.

[0139] This embodiment realizes the interactive visualization of the model through the in-depth application of the BIM rendering engine, improves the dynamic adaptability of the model with the help of the risk level-driven self-optimization mechanism, and builds a data interaction channel with the cloud platform through the standardized export module, forming a complete output system covering visual display, dynamic update and data sharing, providing technical support for the information flow throughout the life cycle of geotechnical engineering.

[0140] Example 9: To address the issues of insufficient stability analysis depth and incomplete consideration of dynamic factors during geotechnical information model reconstruction, this example further optimizes the specific steps of parameter-driven model reconstruction. First, the geotechnical stability analysis algorithm is integrated into the model reconstruction stage, and the limit equilibrium method is used to construct the analysis model. The specific expression is:

[0141] , where K is the stability safety factor, c l is the cohesion of the lth slip surface unit, l l is the length of the slip surface of the unit, σ l is the normal stress acting on the element, ϕ l is the internal friction angle, T l is the total tangential force on the unit slip surface, α l The potential sliding surface is the angle between the slip surface and the horizontal plane. The potential sliding surface is a continuous curved surface where shear failure may occur under the action of the rock mass's own weight or external load. Its determination is the core of stability analysis, used to determine whether the rock mass is in a critical instability state.

[0142] Furthermore, the analysis algorithm searches for potential sliding surfaces that minimize the safety factor K by traversing the three-dimensional grid cells in the model. This process, combined with Monte Carlo simulation, addresses parameter uncertainty and improves the reliability of the analysis results. In the dynamic factor adjustment process, dynamic factors are clearly defined, including groundwater level fluctuations, temporal changes in ground loads, stratum creep effects, and seismic parameter disturbances. For example, groundwater level fluctuations affect effective stress by modifying pore water pressure, changes in ground loads directly alter the normal stress distribution on the slip surface, stratum creep effects adjust the time-dependent attenuation model of cohesion through time parameters, and seismic parameter disturbances introduce inertial force terms to modify the force balance equation of the slip surface.

[0143] Based on real-time monitoring of dynamic factors, the system automatically updates the corresponding parameters in the analysis algorithm and triggers the model reconstruction process. After each parameter adjustment, the algorithm recalculates the location of the potential sliding surface and the safety factor, generating a stability report that includes the spatial coordinates of the sliding surface, a safety factor variation curve, and the response intervals of sensitive parameters. It is important to understand that the report content is linked to the three-dimensional spatial information of the BIM model, allowing for intuitive display of the spatial distribution and evolution of the potential sliding surface in a visual interface.

[0144] This embodiment clarifies the calculation logic of the potential sliding surface by introducing a limit equilibrium analysis algorithm and constructs a parameter adjustment mechanism based on specific dynamic factors. This enables the model reconstruction process to quantitatively reflect the stability state of the rock and soil mass and the interactive relationship between its influencing factors, providing a stability analysis basis based on actual boundary conditions for engineering design.

[0145] Example 10: In order to solve the problems of low system modularization coordination efficiency and unclear data processing logic in the process of constructing a three-dimensional geotechnical engineering survey information model based on BIM, this example further refines the hardware architecture and functional implementation of the construction system. The construction system includes interrelated functional modules, among which the data acquisition module is deployed at the geological survey site. By integrating GNSS positioning equipment, geotechnical mechanics sensors and groundwater monitors, it collects three-dimensional coordinates of geological boreholes, soil density, moisture content, elastic modulus and other attribute samples, as well as groundwater level dynamic data in real time. The acquisition process uses time-division multiplexing technology to synchronize the clock signals of various types of sensors to ensure the temporal consistency of multi-source data.

[0146] Specifically, the preprocessing module is directly connected to the data acquisition module. First, it performs a cleaning operation on the original data and eliminates obvious outliers by setting a value range filter; finally, it converts the unstructured data into a structured data set through a data mapping algorithm, forming a standardized data record containing spatial coordinates, timestamps, and physical parameters.

[0147] The BIM engine module integrates the structured data output by the preprocessing module. It first initializes a 3D spatial reference coordinate system, using the National Geodetic Coordinate System as a benchmark and establishing a unified coordinate framework based on the local elevation benchmark of the project area. It then constructs a model architecture that dynamically integrates stratigraphic information. Using a stratigraphic sedimentation modeling algorithm, it automatically identifies geotechnical interfaces and maps attribute data by stratigraphic level, forming a 3D stratigraphic grid with spatial topological relationships.

[0148] The model reconstruction module interacts with the BIM engine module, and its data fusion submodule receives the dynamic weight vector of geotechnical properties. This vector dynamically adjusts the influence of each parameter on model construction based on the reliability assessment results of the survey data, such as giving higher weight to groundwater level data monitored frequently; the geological constraint submodule integrates the physical constraints of the geotechnical body, including partial differential equation constraints such as the stress equilibrium equation and the seepage continuity equation, to form a set of boundary conditions for model construction; the algorithm processor submodule performs dynamic field integration operations based on the above inputs, and fits the geotechnical property field distribution in three-dimensional space through the finite element interpolation algorithm to generate a continuous parameterized model; the visualization submodule converts the reconstruction results into an intuitive three-dimensional view, supporting real-time browsing and attribute query.

[0149] The output optimization module is connected to the model reconstruction module. First, it performs visualization processing through the BIM rendering engine to achieve multi-dimensional display of the model. Secondly, it establishes a self-optimization feedback channel. When data updates or boundary condition changes are monitored, it triggers the local reconstruction instruction of the model. Finally, the model is converted into a standard format through the data export interface to support interactive applications with other engineering information systems.

[0150] This embodiment builds a complete technical chain from data acquisition to model output by clarifying the hardware configuration and data processing flow of each functional module. The modules work together through standardized interfaces. The sub-module design of the model reconstruction module ensures the dynamic integration of geotechnical properties and the accurate expression of physical constraints, providing a systematic implementation solution for the construction of BIM-based three-dimensional survey information models.

Claims

1. A method for constructing a three-dimensional geotechnical engineering survey information model based on BIM, characterized in that: The steps of the construction method include: Acquire multi-dimensional geotechnical survey data. Collect geotechnical survey data from multiple sources in real time, including geological borehole location coordinates, soil sample properties, groundwater level dynamic values, and geological structure maps. During the collection process, perform data synchronization operations to ensure time stamp consistency. Cleaning the survey data to remove outliers and converting the data format based on a preset geotechnical engineering geology classification standard to generate a unified structured data set; Initialize the multi-dimensional space framework based on the BIM model, start the BIM platform interface, input the unified structured data set, and construct a three-dimensional space reference coordinate system. The three-dimensional space reference coordinate system locates its origin by dynamically integrating geotechnical layer information and spatial geographic coordinates; Parameter-driven geotechnical information model reconstruction, applying parameter-driven mechanism in the three-dimensional spatial reference coordinate system, inputting geotechnical attribute dynamic weight vector, combining geophysical constraints, automatically generating three-dimensional geotechnical engineering investigation information model, including constructing multidimensional kernel function , used to quantize spatial points With reference point The attribute correlation between the two, the multidimensional kernel function It includes two types of core components: Geological anisotropy kernel: The Gaussian kernel is used to characterize the attenuation relationship between lithologic differences and spatial distance. The expression is: , where α is the anisotropic strength coefficient; Stress conduction kernel: The long-range attenuation kernel is used to describe the spatial conduction effect of the tectonic stress field. The expression is: , where β is the stress transmission coefficient, σ1 and σ2 are the spatial correlation scales, and σ2>σ1; The multidimensional kernel function is a weighted sum of two types of core components: ； Construct the property field reconstruction formula including the density field integral term and the thermal-mechanical coupling term: ,in, is the density distribution function, is the normalization factor, is the heat conduction simulation coefficient, is the stress field function, λ is the thermal-mechanical coupling coefficient, and Ω represents the survey area; Define the reconstruction error function: , where φ meas (x i ) is the measured attribute value, λ reg is the regularization coefficient; solving the expression to minimize the error by using the gradient descent method, and outputting a reconstructed geotechnical property field to maintain the dynamic response capability of the geotechnical heterogeneous body model; The survey information model is output, visualized using a rendering engine of a BIM platform, and a self-optimization feedback loop is initiated based on the engineering geological risk level, and the survey information model is applied to a design decision support system.

2. The method for constructing a three-dimensional geotechnical engineering survey information model based on BIM according to claim 1, characterized in that: The steps of generating the unified structured data set include: setting an outlier detection threshold based on geotechnical engineering specifications, performing a filtering operation on the survey data to remove outliers and fill in missing data points; Converting the filtered data into a BIM-compatible format, including encoding soil sample properties into IFC entity objects; Logically assign data labels based on geological hierarchies to generate a geotechnical attribute matrix; Adding timestamps and geological context descriptors to the geotechnical attribute matrix, constructing the unified structured data set, and performing quality assessment operations to calculate the data confidence score. If the score is lower than a preset threshold, reacquiring the survey data.

3. The method for constructing a three-dimensional geotechnical engineering survey information model based on BIM according to claim 1, characterized in that: The steps of constructing the three-dimensional space reference coordinate system include: dynamically calculating the position of the geological reference origin based on the unified structured data set; Divide the multi-layer geotechnical space blocks based on the geological reference origin; Dynamic data fusion factors are applied to adjust block weights to generate an adaptive three-dimensional spatial reference coordinate system, which is then interactively bound to the BIM platform interface to activate a real-time data update module.

4. The method for constructing a three-dimensional geotechnical engineering survey information model based on BIM according to claim 1, wherein: Before outputting the survey information model, a model error correction and feedback mechanism is executed, the steps comprising: Collecting the deviation value between the output point of the survey information model and the actual measured point of the geological drilling; Calculate credibility intervals based on the Bayesian algorithm; Feedback adjustment is initiated based on the credibility interval to optimize the parameter driving mechanism.

5. The method for constructing a three-dimensional geotechnical engineering survey information model based on BIM according to claim 1, wherein: The steps of obtaining the survey data include: The three-dimensional coordinates of multiple geological borehole locations, including longitude, latitude, and depth coordinates, are collected by a sensor array, and rock and soil samples of each borehole are continuously measured to generate an initial data stream; Using a time synchronization protocol to align the initial data stream with a geographic information system layer in real time, and calculating a time delay compensation value to ensure data timing consistency; A cross-validation operation is performed to compare the initial data stream with a preset geotechnical engineering threshold. If any data point exceeds the threshold range, a data re-collection module is triggered to perform redundant collection to output an enhanced multi-dimensional geotechnical engineering investigation dataset.

6. The method for constructing a three-dimensional geotechnical engineering survey information model based on BIM according to claim 1, wherein: The step of outputting the survey information model includes: Apply BIM rendering engines to perform 3D visualization and convert the model into interactive views; Based on the practicality of the engineering geological risk level assessment model, when the risk level is high, a self-optimization cycle is triggered to update the model. At the same time, an output module is added to export the model to a cloud database format.

7. The method for constructing a three-dimensional geotechnical engineering survey information model based on BIM according to claim 1, wherein: The step of reconstructing the parameter-driven geotechnical information model further includes: Execute geotechnical stability analysis algorithms to calculate potential sliding surfaces based on 3D models; Adjust analysis parameters through dynamic factors and generate stability reports.

8. A BIM-based 3D geotechnical engineering investigation information model construction system, used to implement the BIM-based 3D geotechnical engineering investigation information model construction method according to any one of claims 1 to 7, characterized in that: The build system includes: Data acquisition module, used to collect multi-dimensional geotechnical engineering survey data in real time, including geological drill hole coordinates, soil property samples and groundwater level values; a preprocessing module, connected to the data acquisition module, for cleaning, standardizing and transforming the data to generate a structured data set; A BIM engine module, integrating the output of the preprocessing module, for initializing a three-dimensional spatial reference coordinate system and dynamically fusing layer information; A model reconstruction module, interacting with the BIM engine module, for generating a three-dimensional geotechnical engineering investigation information model through a parameter-driven mechanism; an output optimization module, connected to the model reconstruction module, for performing visualization processing, applying self-optimization feedback, and exporting model data; The model reconstruction module includes: Data fusion submodule, used to input geotechnical attribute dynamic weight vector; The geological constraints submodule is used to integrate physical constraints; an algorithm processor submodule, interacting with the data fusion submodule and the geological constraint submodule to perform a dynamic field integration operation; Visualization submodule, used to display the reconstructed model.

Images