Foundation pit excavation deformation prediction method and device based on finite element

Abstract

The invention provides a foundation pit excavation deformation prediction method and device based on finite elements, and relates to the technical field of foundation pit construction.The method comprises the steps that firstly, a digital geologic model is built based on geological survey drilling data, and a foundation pit excavation finite element numerical simulation model is built in combination with a foundation pit supporting design drawing of a design institute; orthogonal design parameters are combined to obtain multiple groups of simulation data, and a twinborn database is established; constructing an initial prediction model and training by using a twin database to obtain a target prediction model; obtaining foundation pit monitoring data through a sensor, extracting data corresponding to the twin database to form an effective parameter combination, and inputting a target model to predict a foundation pit excavation deformation result in the next stage; and judging whether a foundation is excavated or not according to a deformation result, and if not, continuing to predict until the foundation is reached. According to the method, modeling is performed by combining geology and design data, the model is trained by using a twin database, deformation can be accurately predicted, dynamic monitoring and real-time prediction are realized by combining monitoring data, and construction safety is guaranteed.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of foundation pit construction, and particularly relates to a method and device for predicting foundation pit excavation deformation based on finite element. Background Art

[0002] With the acceleration of the modernization process, the utilization of urban above-ground space has approached saturation. To address the problem of urban land resource shortage, the reasonable and efficient utilization of underground space has become an effective solution. The construction of underground projects, such as basements of high-rise buildings, underground shopping malls, underground parking lots, large drainage and sewage treatment systems, underground substations, and large subway stations, all rely on the basic project of foundation pit excavation. While the number of foundation pits is increasing, the potential construction risks are also increasing. As the foundation and key step of underground space development, foundation pit engineering involves many factors, and the construction process has unpredictability and risk.

[0003] Although there have been studies attempting to combine machine learning and finite element methods in recent years, there are still problems such as lack of model training data and real-time feedback mechanism. For example, existing finite element analyses are mostly based on static geological models and fixed parameters, which are difficult to reflect the spatial variability of geological conditions, resulting in prediction lagging behind the actual excavation process and being difficult to guide dynamic construction decisions. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and device for predicting foundation pit excavation deformation based on finite element in view of the deficiencies of the prior art.

[0005] The technical solution of the present invention for solving the above technical problem is as follows: A method for predicting foundation pit excavation deformation based on finite element, comprising: S1. Establish a digital geological model based on pre-collected geological exploration borehole data, and construct a finite element numerical simulation model for foundation pit excavation according to the foundation pit support design drawings of the design institute and in combination with the digital geological model; S2. Conduct orthogonal design parameter combinations on the constructed finite element numerical simulation model for foundation pit excavation to obtain multiple groups of finite element numerical simulation data, and establish a twin database through the multiple groups of finite element numerical simulation data; S3. Construct an initial prediction model, and train the initial prediction model through the established twin database to obtain a target prediction model; S4. Obtain the monitoring data of the foundation pit in real time through pre-deployed sensors, extract the data corresponding to the twin database from the monitoring data and combine them to obtain an effective parameter combination; S5. Input the effective parameter combination as input parameters into the target prediction model to predict the deformation result of the next stage of foundation pit excavation; S6. Determine whether the excavation has reached the foundation base of the foundation pit according to the deformation result. If not, return to S4 to predict the next stage of foundation pit excavation until the foundation base of the target building is reached.

[0006] Another technical solution for the present invention to solve the above technical problems is as follows: A finite element-based foundation pit excavation deformation prediction device includes: A finite element model construction module, configured to establish a digital geological model based on pre-collected geological exploration borehole data, and construct a finite element numerical simulation model for foundation pit excavation according to the foundation pit support design drawings of the design institute and in combination with the digital geological model; A twin database establishment module, configured to perform orthogonal design parameter combinations on the constructed finite element numerical simulation model for foundation pit excavation to obtain multiple groups of finite element numerical simulation data, and establish a twin database through the multiple groups of finite element numerical simulation data; A target prediction model construction module, configured to construct an initial prediction model, and train the initial prediction model through the established twin database to obtain a target prediction model; An effective parameter combination module, configured to obtain the monitoring data of the foundation pit in real time through pre-deployed sensors, extract the data corresponding to the twin database from the monitoring data and combine them to obtain an effective parameter combination; A deformation prediction module, configured to input the effective parameter combination as an input parameter into the target prediction model to predict the deformation result of the next stage of foundation pit excavation; It is also configured to determine whether the excavation has reached the foundation base of the foundation pit according to the deformation result. If not, perform the prediction of the next stage of foundation pit excavation until the foundation base of the target building is reached.

[0007] Another technical solution for the present invention to solve the above technical problems is as follows: A finite element-based foundation pit excavation deformation prediction device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the finite element-based foundation pit excavation deformation prediction method as described above is implemented.

[0008] Another technical solution for the present invention to solve the above technical problems is as follows: A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the finite element-based foundation pit excavation deformation prediction method as described above is implemented.

[0009] The beneficial effects of the present invention are as follows: By combining geological exploration data and foundation pit support design drawings to construct a finite element numerical simulation model, and training a prediction model using a twin database, it is possible to more accurately predict the deformation during the foundation pit excavation process, providing a reliable reference basis for construction. By combining real-time monitoring data with the twin database to generate effective parameter combinations and inputting them into the target prediction model for prediction, dynamic monitoring and real-time prediction of the foundation pit excavation process are achieved, enabling potential problems to be detected in a timely manner and corresponding measures to be taken to ensure construction safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a schematic flowchart of the foundation pit excavation deformation prediction method provided by an embodiment of the present invention; Figure 2 It is an implementation flowchart of the foundation pit excavation deformation prediction method provided by an embodiment of the present invention; Figure 3 It is a functional module block diagram of the foundation pit excavation deformation prediction device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0011] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0012] As Figure 1 shown, an embodiment of the present invention provides a foundation pit excavation deformation prediction method based on finite element, including: S1. Establish a digital geological model based on the pre-collected geological exploration borehole data, and construct a finite element numerical simulation model for foundation pit excavation according to the foundation pit support design drawings of the design institute and in combination with the digital geological model; S2. Perform orthogonal design parameter combinations on the constructed finite element numerical simulation model for foundation pit excavation to obtain multiple groups of finite element numerical simulation data, and establish a twin database through the multiple groups of finite element numerical simulation data; S3. Construct an initial prediction model, and train the initial prediction model through the established twin database to obtain a target prediction model; S4. Obtain the monitoring data of the foundation pit in real time through pre-deployed sensors, extract the data corresponding to the twin database from the monitoring data and combine them to obtain an effective parameter combination; S5. Input the effective parameter combination as input parameters into the target prediction model to predict the deformation result of the next stage of foundation pit excavation; S6. Judge whether the excavation reaches the bottom of the foundation pit according to the deformation result. If not, return to S4 for prediction of the next stage of foundation pit excavation until the bottom of the target building is reached.

[0013] The following describes the entire processing flow from geological exploration to deformation prediction of foundation pit excavation through a specific foundation pit project, as Figure 2 shown below: 1) Geological exploration and digital geological model establishment: Collect geological exploration borehole data; Establish a digital geological model.

[0014] 2) Foundation pit support design: Based on the foundation pit support design drawings of the design institute.

[0015] 3) Construction of finite element numerical simulation model: Construct a finite element numerical simulation model for foundation pit excavation.

[0016] 4) Preliminary calculation and establishment of twin database: Conduct preliminary calculations on the finite element model to obtain preliminary calculation results.

[0017] Conduct orthogonal design parameter combinations.

[0018] Calculate multiple groups of finite element numerical simulation data.

[0019] Establish a twin database.

[0020] 5) Real-time monitoring data storage system: Collect horizontal displacement data of the support structure, soil stress data, water level height data, and environmental parameter data.

[0021] Conduct data preprocessing, data processing, and analysis.

[0022] Extract data combinations corresponding to the twin database.

[0023] 6) Parameter sensitivity analysis and prediction model construction: Conduct sensitivity analysis on input parameters.

[0024] Construct an initial prediction model, determine the input layer and output layer, train the model, and establish the corresponding relationship. 7) Model training and parameter update: Use the new parameter combination values to update the parameters in the initial calculation model.

[0025] Obtain the target prediction model through training.

[0026] 8) Deformation prediction and construction adjustment: Input the effective parameter combination as the input parameter into the target prediction model to predict the deformation result of the next stage of foundation pit excavation.

[0027] Judge whether to excavate to the foundation of the foundation pit according to the deformation result.

[0028] If the foundation base is not reached, return to predict the next stage of foundation pit excavation until the foundation base of the target building is reached.

[0029] In the above embodiments, by combining geological exploration data and foundation pit support design drawings to construct a finite element numerical simulation model and using the twin database to train the prediction model, the deformation situation during the foundation pit excavation can be predicted more accurately, providing a reliable reference basis for construction.

[0030] Combining real-time monitoring data with the twin database to generate effective parameter combinations and inputting them into the target prediction model for prediction realizes the dynamic monitoring and real-time prediction of the foundation pit excavation process, enabling timely discovery of potential problems and taking corresponding measures to ensure construction safety.

[0031] Judge whether it is necessary to adjust the construction plan according to the prediction results. If the predicted deformation exceeds the safety threshold, an early warning can be triggered in advance and support structure reinforcement measures can be recommended, thereby optimizing the construction plan, reducing construction risks, and improving construction efficiency.

[0032] Preferably, in S1, a digital geological model is established based on pre-collected geological exploration borehole data, including: Assume that the collected borehole sample data set is , where is the data value of the i-th borehole sample. Use the Kriging interpolation formula to perform spatial interpolation on the borehole sample data values to establish a digital geological model as: , where is the borehole sample data value at the point to be estimated , is the weight coefficient, is calculated from the variogram and needs to satisfy , Based on the variogram solve for the weight coefficient . The variogram is: , where h is the distance between borehole sample data points, is the number of pairs of borehole sample data points with a spacing of h, and are the borehole sample data values at the spatial positions and in the foundation pit; Generate a stratigraphic distribution map based on the digital geological model imported into the mapping tool, divide the stratigraphic distribution map into j cells according to different strata i, and extract the material properties corresponding to the soil layer of stratum i from the pre-collected geological exploration borehole data. The material properties include elastic modulus , Poisson's ratio , internal friction angle and cohesion . Calculate the elastic modulus , Poisson's ratio , internal friction angle and cohesion corresponding to the j cells respectively through the weighted average method. For the elastic modulus corresponding to the j-th cell it is: . For the Poisson's ratio corresponding to the j-th cell it is: . For the internal friction angle corresponding to the j-th cell it is: . For the cohesion corresponding to the j-th cell it is: . where n is the number of boreholes adjacent to the j-th cell, is the weight coefficient corresponding to the elastic modulus, is the weight coefficient corresponding to the Poisson's ratio, is the weight coefficient corresponding to the internal friction angle, is the weight coefficient corresponding to the cohesion.

[0033] It should be understood that Kriging interpolation is a geostatistical method that can estimate the geological property values at unknown locations based on the known borehole sample data. It takes into account the spatial autocorrelation of the data, that is, the closer the points are, the more similar their geological properties are. In this way, a continuous geological property distribution map can be generated to more accurately reflect the spatial variation of the geological conditions.

[0034] Variogram The variogram is the core of Kriging interpolation, which describes the spatial correlation of geological property values with distance. Through the variogram, the weight coefficients can be calculated to determine the contribution degree of each borehole sample data to the estimation point. The calculation result of the variogram directly affects the distribution of the weight coefficients. The weight coefficients determine the importance of each borehole sample data in the interpolation process. A reasonable distribution of weight coefficients can improve the accuracy of interpolation and avoid estimation biases caused by uneven data distribution.

[0035] When calculating the material properties of each cell, the weighted average method is adopted, which can fully consider the weights of the borehole data adjacent to the cell. The weight coefficients can be adjusted according to factors such as the distance between the borehole and the cell and the reliability of the data, making the calculation result more representative. By using the weighted average method, the errors that may be caused by simple averaging can be avoided. For example, if a cell is close to multiple boreholes and the data of these boreholes vary greatly, simple averaging may lead to large errors. The weighted average method can reasonably calculate the material property value of the cell according to the weights of each borehole data, thereby improving the accuracy of the entire digital geological model.

[0036] To convert the digital geological model into a visual stratigraphic distribution map, appropriate mapping tools need to be selected. Common mapping tools include GIS (Geographic Information System) tools, geological modeling tools, and general 3D modeling tools.

[0037] For example, use the ArcGIS tool to generate the stratigraphic distribution map: Data preparation: Export the data of the digital geological model as a CSV file, including the coordinates (X, Y, Z) of each point and the geological property values (such as the elastic modulus E).

[0038] Data import: In ArcGIS, use the "Add Data" function to load the CSV file.

[0039] Grid division: Use the "Create Fishnet" tool to divide the foundation pit area into grids.

[0040] Attribute assignment: Use the "Spatial Interpolation" tool (such as Kriging interpolation) to assign geological property values to each grid cell.

[0041] Visualization settings: Select the "Symbol System" option, set the color mapping, and represent different elastic modulus values with different colors.

[0042] Map generation: The finally generated stratigraphic distribution map can be exported as a PDF or other format for further analysis or display.

[0043] Through this process, the digital geological model is effectively transformed into an intuitive stratigraphic distribution map, providing important visualization support for the prediction of foundation pit excavation deformation and construction management.

[0044] In the above embodiment, the Kriging interpolation formula is used to perform spatial interpolation on the borehole sample data, which can more accurately establish the digital geological model, fully consider the spatial variability and complexity of geological conditions, and provide a more accurate geological basis for subsequent finite element numerical simulation.

[0045] The stratigraphic distribution map is divided into multiple cells, and the material properties of each cell are calculated by the weighted average method, enabling the finite element model to more precisely reflect the mechanical properties of different strata, improving the accuracy of finite element numerical simulation, and further enhancing the reliability of the prediction results.

[0046] Preferably, in S1, based on the foundation pit support design drawings provided by the design institute and in combination with the digital geological model, a finite element numerical simulation model for foundation pit excavation is constructed, including: Drawing the foundation pit geometric model by a drawing tool according to the foundation pit support design drawings provided by the design institute, mapping the foundation pit geometric model to the stratigraphic distribution map, and dividing the foundation pit geometric model into finite element mesh units corresponding to j cells in the stratigraphic distribution map, and assigning the material properties of each finite element mesh unit as the elastic modulus of the corresponding cell , Poisson's ratio , internal friction angle and cohesion ; Establishing the material stiffness matrix D corresponding to the jth finite element mesh unit and defining the plastic behavior of the soil mass, specifically: For a linearly elastic material under plane strain conditions, its material stiffness matrix D is expressed as: , And using the Mohr-Coulomb yield function to define the plastic behavior of the soil mass, the Mohr-Coulomb yield function is: , where is the first stress invariant, is the second deviatoric stress invariant, to obtain the finite element numerical simulation model for foundation pit excavation.

[0047] It should be understood that the foundation pit support design drawings provided by the design institute contain detailed information such as the shape, size, position and type of the support structure of the foundation pit. These information are the basis for constructing the finite element geometric model.

[0048] Use professional drawing software (such as AutoCAD, GeoStudio, etc.) to draw the geometric shape of the foundation pit according to the design drawings. This includes the boundaries of the foundation pit and the positions and shapes of the support structures (such as retaining walls, anchor bolts, etc.). Align and map the drawn geometric model of the foundation pit to the stratum distribution map in the digital geological model. Ensure that each part of the geometric model of the foundation pit accurately corresponds to the corresponding stratum cells. This step is crucial for subsequent mesh generation and material property assignment.

[0049] The mesh generation method is the finite element method, which requires dividing the complex geometric model into small elements (meshes) for numerical calculations. The quality of the mesh generation directly affects the accuracy and efficiency of the finite element analysis.

[0050] According to the geometric shape of the foundation pit and the stratum distribution, select a suitable mesh type (such as quadrilateral mesh, triangular mesh, etc.). For complex geometric shapes, a hybrid mesh may be required. In key areas (such as near the support structure and areas with drastic stratum changes), a denser mesh is needed to improve the calculation accuracy. In relatively flat areas, a sparser mesh can be used to reduce the amount of calculation.

[0051] Assign the material properties of each finite element mesh element to the property values of the corresponding cell. This can be done by interpolation or direct assignment. For example, if a mesh element is completely within a certain stratum cell, directly assign the property value of that stratum; if the mesh element spans multiple stratum cells, its property value needs to be calculated by methods such as weighted averaging.

[0052] In the above embodiments, the geometric model of the foundation pit is drawn based on the foundation pit support design drawings of the design institute and mapped to the stratum distribution map, which can accurately simulate the actual geometric shape and positional relationship of the foundation pit, providing an accurate geometric basis for the finite element numerical simulation.

[0053] Establish the material stiffness matrix of the finite element mesh elements and use the Mohr - Coulomb yield function to define the plastic behavior of the soil, which can more reasonably describe the mechanical response of the soil during the excavation process, making the finite element numerical simulation model closer to the actual engineering situation and improving the accuracy of the prediction results.

[0054] Preferably, in S2, perform an orthogonal design parameter combination on the constructed finite element numerical simulation model of the foundation pit excavation to obtain multiple groups of finite element numerical simulation data, and establish a twin database through the multiple groups of finite element numerical simulation data. Specifically: Based on each finite element mesh element in the finite element numerical simulation model of the foundation pit excavation, determine the set of input parameters to be optimized , and divide k horizontal constraint ranges for each parameter according to the geological exploration data: , , , Among them, 、 、 and are determined by geological exploration data respectively; Using the orthogonal experimental design method, select the orthogonal table , generate N groups of parameter combinations Among them, is the number of parameters, k is the number of levels, ; For each group of parameter combinations , extract the key node displacement data and stress data based on the finite element numerical simulation model of foundation pit excavation, and form an output data set ; Associate and store the parameter combinations and the output data set to build a twin database .

[0055] It should be understood that, based on the geological exploration data, a reasonable horizontal constraint range is divided for each parameter. For example, the elastic modulus may vary within a certain range, and the Poisson's ratio also has its reasonable value range.

[0056] Using the orthogonal experimental design method, select a suitable orthogonal table and generate multiple groups of parameter combinations. The orthogonal experimental design is an efficient experimental design method that can comprehensively investigate the interaction of multiple factors with fewer experimental times.

[0057] The twin database contains simulation results under various parameter combinations, provides rich training samples for the machine learning model, and helps to improve the generalization ability of the model.

[0058] In the above embodiments, by determining the input parameter set to be optimized and dividing the horizontal constraint range, and then using the orthogonal experimental design method to generate multiple groups of parameter combinations, a large amount of finite element numerical simulation data can be efficiently generated, and a twin database covering various working conditions can be constructed, providing rich data support for subsequent model training, and helping to improve the generalization ability and prediction accuracy of the prediction model.

[0059] Associating and storing the parameter combinations and the output data set to build a twin database realizes the effective organization and management of data, facilitates subsequent data query, analysis and model training, and improves the operation efficiency of the entire prediction system.

[0060] Preferably, in S3, an initial prediction model is constructed, and the initial prediction model is trained through the established twin database to obtain a target prediction model, including: For the parameter combination the data in it and the key node displacement data the data in it are normalized respectively as: , , where and are the mean and standard deviation of the parameter combination respectively, and and are the mean and standard deviation of the key node displacement data respectively; An initial prediction model is constructed based on the PSO - BP neural network. The initial prediction model includes an input layer, a hidden layer, and an output layer, and the number of nodes in the input layer and the output layer is set, as well as the activation function, the number of layers, and the number of nodes in each layer of the hidden layer are set, and the training parameters are defined, and the initial weights of the initial prediction model are optimized through the particle swarm PSO algorithm; The initial prediction model is trained through the normalized training data set and the weights and bias parameters of the initial prediction model are updated through the backpropagation algorithm to obtain a target prediction model.

[0061] It should be understood that the particle swarm optimization (PSO) algorithm is an optimization algorithm based on swarm intelligence, which finds the optimal solution by simulating the foraging behavior of bird flocks. In a neural network, the PSO algorithm can be used to optimize the initial weights of the network to avoid the problem that the traditional BP neural network is prone to falling into local optima.

[0062] The initial prediction model includes an input layer, a hidden layer, and an output layer. The number of nodes in the input layer is usually the same as the number of input parameters, and the number of nodes in the output layer is the same as the number of prediction targets. The number of nodes and the number of layers of the hidden layer need to be adjusted according to specific problems to achieve the best prediction effect.

[0063] For example: Input layer: The number of nodes is 4, corresponding to the parameter combination; Hidden layer: It contains 2 layers, the number of nodes in each layer is 10, the activation function is ReLU, which is used to capture the non - linear relationship between the input and the output; Output layer: The number of nodes is 3, corresponding to the key node displacement data, the activation function is linear, and the standardized displacement prediction value is output.

[0064] The defined training parameters include: the loss function is the mean squared error (MSE), which is used to measure the deviation between the predicted value and the true value; the optimizer is the Adam algorithm, with the learning rate set to 0.001 to accelerate the model convergence; the number of training epochs is 500, and the batch size is 32 to ensure sufficient data learning.

[0065] Apply the Particle Swarm Optimization (PSO) algorithm to optimize the initial weights of the neural network, specifically including: Particle swarm parameters: population size 5, maximum number of iterations 50; Optimization objective: minimize the fitness value of the initial weights of the neural network, and the fitness function is the MSE on the training set; Weight update rule: adjust the particle position through the weighted combination of the global optimum and the individual optimum.

[0066] In the above embodiments, normalizing the parameter combinations and the key node displacement data can eliminate the influence of data dimension and order of magnitude differences on model training, making the model training more stable and efficient. Constructing an initial prediction model based on the PSO-BP neural network and optimizing the initial weights through the PSO algorithm can effectively avoid the problem that the traditional BP neural network is prone to falling into local optima, improve the convergence speed and prediction accuracy of the model, and thus obtain a more accurate target prediction model.

[0067] Preferably, in S4, the monitoring data of the foundation pit is obtained in real time through pre-deployed sensors, and the data corresponding to the twin database is extracted from the monitoring data and combined with it to obtain effective parameter combinations, including: The monitoring data of the foundation pit is obtained in real time through pre-deployed sensors, and the monitoring data includes the horizontal displacement data of the supporting structure obtained by displacement sensors, the soil stress data obtained by earth pressure cells, the water level height data obtained by water level sensors, and the environmental parameters obtained by temperature and humidity sensors; Extract the feature vectors of the horizontal displacement data of the supporting structure, the soil stress data, and the water level height data corresponding to the twin database from the monitoring data to obtain real-time feature vectors , where is the displacement feature, is the stress feature, is the water level feature; Calculate the similarity between the real-time feature vector and the feature vector corresponding to each parameter combination in the twin database : , where is the similarity distance between the l-th group of parameter combinations and the real-time feature vector. The smaller the value, the more similar the real-time data is to the features of the l-th group of parameters in the twin database. is the actual measured value of the i-th feature in the real-time feature vector. is the simulated value of the i-th feature corresponding to the l-th group of parameter combinations in the twin database. is the value range of the i-th dimension feature in the twin database. Select the top K groups of parameter combinations with the highest similarity ; Perform weighted fusion on the selected parameter combinations to generate effective parameter combinations : , , where the weight is inversely proportional to the similarity to ensure that parameter combinations with high similarity contribute more to the fusion result. is the similarity distance of the k-th group among the top K groups of parameter combinations after screening. , According to the k-level constraint ranges, truncate the values in the fused effective parameter combination that exceed the k-level constraint ranges to the boundary values.

[0068] In the above embodiments, by using a variety of pre-deployed sensors to obtain the monitoring data of the foundation pit in real time and extracting the feature vectors corresponding to the twin database, the on-site real-time monitoring data can be fully utilized to provide the latest engineering information for the prediction model, enhancing the timeliness and accuracy of the prediction results. By calculating the similarity between the real-time feature vector and the parameter combinations in the twin database based on the Euclidean distance formula and selecting the top K groups of parameter combinations with the highest similarity for weighted fusion, the parameter combinations closest to the current working conditions can be screened out, generating more practical effective parameter combinations and further improving the reliability of the prediction results.

[0069] Preferably, in step S5, input the effective parameter combination as the input parameter into the target prediction model to predict the deformation result of the foundation pit excavation in the next stage, including: Perform normalization processing on the effective parameter combination ( ); Input the normalized effective parameter combination into the target prediction model, and obtain the standardized displacement prediction value through forward propagation calculation. The calculation formula is: , Among them, the mean and standard deviation of the output displacement of the training data set are obtained; The standardized predicted values are de-normalized to restore them to actual engineering units: , The predicted displacement is matched with the key node coordinates of the finite element model to generate a visualized deformation nephogram and output the maximum displacement value and the displacement change trend curve; According to the maximum displacement value it is judged whether the construction plan needs to be adjusted. If , is the safety threshold allowed by the design, a warning is triggered and support structure reinforcement measures are recommended.

[0070] In the above embodiments, after the effective parameter combinations are normalized and input into the target prediction model, the standardized displacement prediction values are obtained through forward propagation calculation, and then de-normalized to restore them to actual engineering units, which can accurately predict the deformation results of the next stage of foundation pit excavation and provide strong support for construction decision-making.

[0071] Generating a visualized deformation nephogram and outputting the maximum displacement value and the displacement change trend curve can intuitively display the foundation pit deformation situation, facilitating construction personnel to quickly understand the project status. At the same time, judging whether the construction plan needs to be adjusted according to the maximum displacement value, and triggering a warning and recommending reinforcement measures when necessary effectively guarantees the construction safety.

[0072] Preferably, in S6, judging whether the excavation reaches the base of the foundation pit according to the deformation result includes: Calculating the spatial relative displacement deviation between the deformation result and the designed elevation of the base; Judging whether the convergence condition is satisfied according to the spatial relative displacement deviation. If it is satisfied, it is judged that the excavation has reached the base.

[0073] For example, the convergence conditions include: Convergence condition one: , where is the absolute value of the spatial relative displacement deviation, is the engineering allowable error; Convergence condition two: The displacement change rate in z consecutive excavation stages , where is the rate threshold. For example, z = 3, such as 0.1 mm / day.

[0074] In the above embodiments, the spatial relative displacement deviation from the designed elevation of the foundation base is calculated based on the predicted deformation result, and it is determined whether to excavate to the foundation base according to the convergence condition, which can accurately determine whether the foundation pit excavation reaches the foundation base of the target building, avoid over-excavation or under-excavation, and ensure the construction quality. Setting various convergence conditions including the absolute value of the spatial relative displacement deviation and the displacement change rate in the continuous excavation stage can more comprehensively evaluate the convergence of the foundation pit excavation, improve the accuracy and reliability of the judgment, and provide a scientific basis for the control of the construction progress.

[0075] As Figure 3 shown, another embodiment of the present invention provides a finite element-based foundation pit excavation deformation prediction device, including: A finite element model construction module for establishing a digital geological model based on the pre-collected geological exploration borehole data, and constructing a finite element numerical simulation model for foundation pit excavation according to the foundation pit support design drawings of the design institute and in combination with the digital geological model; A twin database establishment module for performing orthogonal design parameter combinations on the constructed finite element numerical simulation model for foundation pit excavation to obtain multiple groups of finite element numerical simulation data, and establishing a twin database through the multiple groups of finite element numerical simulation data; A target prediction model construction module for constructing an initial prediction model and training the initial prediction model through the established twin database to obtain a target prediction model; An effective parameter combination module for obtaining the monitoring data of the foundation pit in real time through pre-deployed sensors, extracting the data corresponding to the twin database from the monitoring data and combining them to obtain an effective parameter combination; A deformation prediction module for inputting the effective parameter combination as input parameters into the target prediction model to predict the deformation result of the next stage of foundation pit excavation; It is also used to judge whether to excavate to the foundation base of the foundation pit according to the deformation result. If not, the prediction of the next stage of foundation pit excavation is carried out until the foundation base of the target building is reached.

[0076] Another embodiment of the present invention provides a finite element-based foundation pit excavation deformation prediction device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the finite element-based foundation pit excavation deformation prediction method as described above is implemented.

[0077] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program, which when executed by a processor, implements the finite element-based foundation pit excavation deformation prediction method as described above.

[0078] Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working processes of the systems and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0079] In several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0080] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments of the present invention.

[0081] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for predicting foundation pit excavation deformation based on finite element, characterized in that: include: S1. A digital geological model is established based on the pre-collected geological survey drilling data. A finite element numerical simulation model of foundation pit excavation is constructed based on the foundation pit support design drawings of the design institute and combined with the digital geological model. S2. Perform orthogonal design parameter combination on the constructed foundation pit excavation finite element numerical simulation model to obtain multiple sets of finite element numerical simulation data, and establish a twin database through the multiple sets of finite element numerical simulation data; S3, construct an initial prediction model, train the initial prediction model through the established twin database, and obtain the target prediction model; S4. Obtaining monitoring data of the foundation pit in real time through pre-deployed sensors, extracting data corresponding to the twin database from the monitoring data and combining them with the twin database to obtain an effective parameter combination; S5, inputting the effective parameter combination as an input parameter into the target prediction model to predict the deformation result of the foundation pit excavation in the next stage; S6. Determine whether the excavation has reached the base of the foundation pit according to the deformation result. If not, return to S4 to predict the next stage of foundation pit excavation until the base of the target building is reached.

2. The method for predicting foundation pit excavation deformation according to claim 1, characterized in that: In S1, a digital geological model is established based on the pre-collected geological survey drilling data, including: Assume that the collected borehole sample data set is ,in, is the i-th borehole sample data value. The Kriging interpolation formula is used to perform spatial interpolation on the borehole sample data value to establish a digital geological model: ， in, Point to be estimated The drill hole sample data value at is the weight coefficient, By the variogram Calculated and must satisfy , Based on the variogram Weight coefficient Solve, the variation function is: ， Where h is the distance between the drilling sample data points, is the number of pairs of drilling sample data points with a spacing of h, and The spatial position in the foundation pit and The drill hole sample data value at ; Based on the digital geological model, a stratigraphic distribution map is generated by importing it into a mapping tool, and the stratigraphic distribution map is divided into j cells according to different stratigraphic layers i, and the material properties corresponding to the i-th soil layer are extracted from the pre-collected geological survey drilling data, and the material properties include elastic modulus , Poisson's ratio , internal friction angle and cohesion , and calculate the elastic modulus corresponding to j cells by weighted average method , Poisson's ratio , internal friction angle and cohesion , For the jth unit cell, the elastic modulus for: ， For the jth cell, the Poisson's ratio for: ， For the internal friction angle corresponding to the jth unit cell for: ， For the cohesion corresponding to the jth cell for: ， Where n is the number of holes adjacent to the jth cell, is the weight coefficient corresponding to the elastic modulus, is the weight coefficient corresponding to Poisson’s ratio, is the weight coefficient corresponding to the internal friction angle, is the weight coefficient corresponding to the cohesion.

3. The method for predicting foundation pit excavation deformation according to claim 2, characterized in that: In S1, a finite element numerical simulation model of foundation pit excavation is constructed based on the foundation pit support design drawings of the design institute and combined with the digital geological model, including: According to the foundation pit support design drawings of the design institute, a foundation pit geometric model is drawn using a drawing tool, the foundation pit geometric model is mapped to the stratigraphic distribution map, and the foundation pit geometric model is divided into finite element grid cells corresponding to j cells in the stratigraphic distribution map, and the material property of each finite element grid cell is assigned to the elastic modulus of the corresponding cell. , Poisson's ratio , internal friction angle and cohesion ; Establish the material stiffness matrix D corresponding to the jth finite element mesh unit and define the plastic behavior of the soil, specifically: For linear elastic materials under plane strain conditions, the material stiffness matrix D is expressed as: ， The Mohr-Coulomb yield function is used to define the plastic behavior of the soil. The Mohr-Coulomb yield function is: ， in, is the first stress invariant, is the second deviatoric stress invariant, The finite element numerical simulation model of foundation pit excavation is obtained.

4. The method for predicting foundation pit excavation deformation according to claim 1, characterized in that: In S2, orthogonal design parameter combination is performed on the constructed foundation pit excavation finite element numerical simulation model to obtain multiple sets of finite element numerical simulation data, and a twin database is established through the multiple sets of finite element numerical simulation data, specifically: Based on each finite element mesh unit in the finite element numerical simulation model of foundation pit excavation, determine the input parameter set to be optimized , and divide each parameter into k level constraint ranges based on geological survey data: ， ， ， ， in, , , and Determined by geological survey data respectively; Use orthogonal experimental design method and select orthogonal table , generate N sets of parameter combinations ,in, is the number of parameters, k is the number of levels, ; For each parameter combination , extracting displacement data of key nodes based on finite element numerical simulation model of foundation pit excavation and stress data , forming the output data set ; Combining parameters and output dataset Associative storage, building a twin database .

5. The method for predicting foundation pit excavation deformation according to claim 4, characterized in that: In S3, an initial prediction model is constructed, and the initial prediction model is trained through the established twin database to obtain a target prediction model, including: Combination of parameters Data in and displacement data of key nodes The data in are normalized as follows: ， ， in, and The parameter combinations are The mean and standard deviation of and They are the displacement data of key nodes. The mean and standard deviation of An initial prediction model is constructed based on a PSO-BP neural network, wherein the initial prediction model includes an input layer, a hidden layer, and an output layer, and the number of nodes of the input layer and the output layer is set, as well as the activation function and the number of layers of the hidden layer and the number of nodes per layer are set, and training parameters are defined, and the initial weight of the initial prediction model is optimized by a particle swarm PSO algorithm; The initial prediction model is trained by using the normalized training data set, and the weight and bias parameters of the initial prediction model are updated by a back propagation algorithm to obtain a target prediction model.

6. The method for predicting foundation pit excavation deformation according to claim 5, characterized in that: In S4, the monitoring data of the foundation pit is obtained in real time through the pre-deployed sensors, and the data corresponding to the twin database is extracted from the monitoring data and combined with the twin database to obtain an effective parameter combination, including: The monitoring data of the foundation pit is obtained in real time through the pre-deployed sensors, and the monitoring data includes the horizontal displacement data of the support structure obtained by the displacement sensor, the soil stress data obtained by the soil pressure box, the water level height data obtained by the water level sensor, and the environmental parameters obtained by the temperature and humidity sensor; Extract the characteristic vectors of the horizontal displacement data of the supporting structure, the soil stress data and the water level height data corresponding to the twin database from the monitoring data to obtain the real-time characteristic vector ,in, is the displacement characteristic, is the stress characteristic, is the water level characteristic; Calculate real-time feature vector based on Euclidean distance formula The feature vector corresponding to each parameter combination in the twin database Similarity: ， in, is the similarity distance between the lth group of parameter combinations and the real-time feature vector, is the actual measured value of the i-th feature in the real-time feature vector, is the simulated value of the ith feature corresponding to the lth parameter combination in the database, is the value range of the i-th dimension feature in the twin database; Select the top K parameter combinations with the highest similarity ; Perform weighted fusion on the selected parameter combinations to generate effective parameter combinations : ， ， Among them, the weight Similarity Inversely proportional, is the similarity distance of the kth group among the first K groups of parameter combinations after screening, , Combine the fused effective parameters according to the k level constraint ranges Values ​​in that exceed the k level constraints are truncated to the boundary values.

7. The method for predicting foundation pit excavation deformation according to claim 5, characterized in that: In S5, the effective parameter combination is input as an input parameter into the target prediction model to predict the deformation result of the next stage of foundation pit excavation, including: For valid parameter combinations Perform normalization processing; The normalized effective parameter combination Input the target prediction model and obtain the normalized displacement prediction value through forward propagation calculation , the calculation formula is: ， Among them, the training data set outputs the mean and standard deviation of the displacement; Denormalize the standardized predicted values ​​and restore them to actual engineering units: ， The predicted displacement Match the key node coordinates of the finite element model, generate a visual deformation cloud map, and output the maximum displacement value And displacement change trend curve; According to the maximum displacement Determine whether the construction plan needs to be adjusted. , If the safety threshold is higher than the design limit, an early warning is triggered and reinforcement measures for the supporting structure are recommended.

8. The method for predicting foundation pit excavation deformation according to claim 7, characterized in that: In S6, judging whether to excavate to the base of the foundation pit according to the deformation result includes: Calculate the spatial relative displacement deviation between the base design elevation and the base design elevation according to the deformation result; Whether the convergence condition is met is determined based on the spatial relative displacement deviation. If so, it is determined that the excavation has reached the base.

9. The method for predicting foundation pit excavation deformation according to claim 8, characterized in that: The convergence conditions include: Convergence condition 1: ,in, is the absolute value of the relative displacement deviation in space, is the engineering allowable error; Convergence condition 2: displacement change rate of z consecutive excavation stages ,in, is the rate threshold.

10. A finite element based foundation pit excavation deformation prediction device, characterized in that: include: The finite element model building module is used to build a digital geological model based on the pre-collected geological survey drilling data, and to build a finite element numerical simulation model for foundation pit excavation based on the foundation pit support design drawings of the design institute and combined with the digital geological model; A twin database establishment module is used to perform orthogonal design parameter combination on the constructed foundation pit excavation finite element numerical simulation model to obtain multiple sets of finite element numerical simulation data, and establish a twin database through multiple sets of finite element numerical simulation data; The target prediction model construction module is used to construct an initial prediction model, train the initial prediction model through the established twin database, and obtain the target prediction model; An effective parameter combination module is used to obtain the monitoring data of the foundation pit in real time through pre-deployed sensors, extract the data corresponding to the twin database from the monitoring data and combine it with the twin database to obtain an effective parameter combination; A deformation prediction module, used to input the effective parameter combination as an input parameter into the target prediction model, and predict the deformation result of the next stage of foundation pit excavation; It is also used to determine whether the excavation has reached the base of the foundation pit according to the deformation result. If not, the next stage of foundation pit excavation is predicted until the base of the target building is reached.

Images