Foundation pit simulation analysis method, system and equipment based on finite element method and medium

Abstract

The invention relates to the technical field of civil engineering, and discloses a foundation pit simulation analysis method based on a finite element method. Comprising the following steps: S1, collecting foundation pit engineering data and environment information: collecting and storing real-time monitoring data of geological conditions, soil mechanical properties, underground water level, climate change and settlement of foundation pit engineering by using geological exploration equipment, an environment monitoring sensor, climate monitoring equipment, an underground water level sensor and a settlement monitoring sensor; and S2, establishing a multi-physics field coupling model of the foundation pit: according to the foundation pit engineering data collected in the step S1, providing a foundation pit simulation analysis system based on the finite element method, foundation pit simulation analysis equipment based on the finite element method and a computer readable storage medium. According to the method, by combining the finite element method, the multi-physics field coupling model and deep learning optimization, efficient and accurate foundation pit stability analysis is achieved, optimization suggestions are automatically generated, and the calculation efficiency, accuracy and real-time dynamic adjustment capacity are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of civil engineering, and specifically to a foundation pit simulation analysis method, system, device and medium based on the finite element method. Background Art

[0002] In modern urban construction, foundation pit excavation is an important part of many underground projects (such as subways, underground parking lots, underground pipelines, etc.). The stability of the foundation pit is directly related to construction safety and project quality. Therefore, accurately evaluating the stability of the foundation pit has become an important part that cannot be ignored during the construction process.

[0003] Currently, the stability analysis of foundation pits usually uses the finite element method for numerical simulation, with the help of professional finite element analysis software. These software can accurately simulate physical phenomena such as soil stress, displacement, and settlement. The finite element analysis methods in the prior art, especially in commercial analysis software (such as ANSYS, ABAQUS), have been widely used in foundation pit design and construction. These software can deeply analyze the soil mechanical behavior and structural stability under complex geological conditions by establishing detailed mathematical models, providing effective tool support for engineering personnel, and greatly improving the safety and accuracy of civil engineering projects. Especially under relatively simple geological conditions, these software can better predict the stability of the foundation pit and provide feasible technical guidance for construction.

[0004] However, the existing finite element analysis software still has some deficiencies in practical applications; firstly, these software usually require high-performance computers for long-term operations, consuming a large amount of computing resources and unable to meet the requirements of projects with high real-time requirements; secondly, although the finite element method can handle complex geological conditions, in the context of multi-physical field coupling, it often relies on simplified models or assumptions, which affects the accuracy and reliability of the simulation results; in addition, the existing finite element analysis tools are complex to operate and have a high technical threshold. Ordinary engineering personnel are difficult to operate proficiently without in-depth study, resulting in difficulties in practical applications; furthermore, the existing technologies mostly rely on static data for analysis and lack the ability to provide real-time feedback and optimization for the dynamic changes during the construction process, and cannot adapt to the changes in on-site conditions in a timely manner. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides a foundation pit simulation analysis method, system, device and medium based on the finite element method, which solves the problems of waste of computing resources, high technical threshold, lack of multi-physical field coupling and real-time dynamic optimization in the prior art.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A foundation pit simulation analysis method based on the finite element method, including the following steps: S1. Collect foundation pit engineering data and environmental information: Use geological exploration equipment, environmental monitoring sensors, climate monitoring equipment, groundwater level sensors, and settlement monitoring sensors to collect and store real-time monitoring data on the geological conditions, soil mechanical properties, groundwater level, climate change, and settlement of the foundation pit project. S2. Establish a multi-physical field coupling model for the foundation pit: Based on the foundation pit engineering data collected in step S1, construct a multi-physical field coupling model for the foundation pit. The model includes the interaction of soil mechanics, groundwater flow, and heat conduction, forming a coupled equation set for foundation pit stability analysis. S3. Conduct finite element analysis based on the multi-physical field coupling model: On the basis of the multi-physical field coupling model constructed in step S2, use the finite element method to simulate the mechanical behavior, groundwater flow, and thermal effects of the foundation pit, and obtain stress, displacement, and settlement analysis results. S4. Conduct non-linear finite element analysis and adaptive mesh optimization: According to the finite element analysis results in step S3, further use a non-linear constitutive model to accurately model the soil, and use an error estimation method to adaptively optimize the mesh. S5. Optimize the simulation model parameters and conduct deep learning optimization: Based on the finite element analysis results described in step S4, use a convolutional neural network to learn historical data and real-time monitoring data, and automatically optimize the parameter settings of the simulation model to enhance the simulation accuracy. S6. Conduct cloud computing parallel computing optimization: On the basis of the optimized model obtained in step S5, perform parallel computing through a cloud computing platform, and distribute the simulation tasks to multiple computing nodes to ensure the efficient processing of large-scale computing tasks. S7. Output a foundation pit safety assessment report and provide optimization suggestions: According to the simulation results obtained in step S6, generate a foundation pit safety assessment report, and provide optimization suggestions based on the safety assessment results.

[0007] The present invention provides a foundation pit simulation analysis method, system, device, and medium based on the finite element method. It has the following beneficial effects: 1. Through the cloud computing platform and parallel computing technology, the present invention ensures the efficient distribution of foundation pit simulation tasks among multiple computing nodes. Different from traditional single-node computing, the system dynamically adjusts resource allocation, improving the computing speed and accuracy. This not only speeds up the large-scale computing process but also effectively solves the problems of resource waste and computing bottlenecks in the prior art.

[0008] 2. By combining the finite element method with the multi-physical field coupling model, the present invention can comprehensively consider multiple factors such as stress, displacement, groundwater flow, and thermal effects, providing a more accurate foundation pit stability analysis. Compared with the prior art solutions that only focus on a single factor, the present invention solves the problem of being unable to accurately evaluate the foundation pit risk in a complex construction environment, ensuring a more reliable construction plan.

[0009] 3. By combining real-time monitoring data with simulation results, the system can dynamically adjust the analysis model and optimization scheme. Different from the prior art that relies on static data, the present invention can reflect various changes in foundation pit construction in real time, improving the ability to respond to emergencies. This innovative method effectively reduces risks and optimizes the construction decision-making process. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a flowchart of the method of the present invention; Figure 2 is a structural diagram of the system of the present invention; Figure 3 is a module architecture diagram of the equipment of the present invention; Figure 4 is a schematic diagram of the storage medium of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0011] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the specification of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0012] Please refer to the attached Figure 1 , the embodiment of the present invention provides a foundation pit simulation analysis method based on the finite element method, including the following steps: S1. Collect foundation pit engineering data and environmental information: Use geological exploration equipment, environmental monitoring sensors, climate monitoring equipment, groundwater level sensors, and settlement monitoring sensors to collect and store real-time monitoring data of the geological conditions, soil mechanical properties, groundwater level, climate change, and settlement of the foundation pit project; Through accurate real-time data collection in S1, not only can it provide the necessary input for subsequent simulation analysis, but also help to monitor the environmental changes of the foundation pit in real time to ensure that the analysis model conforms to the actual situation. The implementation of step S1 provides data support for the successful implementation of subsequent steps such as multi-physical field coupling analysis, nonlinear finite element analysis, and deep learning optimization.

[0013] First of all, in this embodiment, real-time data of the foundation pit project is collected through various technical means such as geological exploration equipment, environmental monitoring sensors, climate monitoring equipment, groundwater level sensors, and settlement monitoring sensors. The data collected by the sensors and equipment will be stored and processed in real time. The data includes but is not limited to: soil mechanical properties (such as elastic modulus, friction angle, etc.), groundwater level, temperature change, settlement situation, etc. These data serve as the basis for subsequent model construction and simulation analysis.

[0014] The data is collected through the following types of sensors and devices: Geological exploration equipment: These devices are used to detect the distribution of soil layers in the foundation pit, the physical properties of each soil layer (such as shear strength, friction angle, elastic modulus, etc.) and underground rock formation information. Commonly used geological exploration tools include drilling equipment, reflection wave measurement equipment, etc. Through a detailed investigation of the foundation pit geological environment, the mechanical properties of the soil at the bottom and side walls of the foundation pit can be obtained, providing physical parameter support for subsequent analysis.

[0015] Environmental monitoring sensors: Such sensors are mainly used to monitor the climate change of the foundation pit construction environment in real time. By collecting environmental data, potential threats to the safety of the foundation pit caused by weather changes can be predicted.

[0016] Underground water level sensors: The change of the underground water level has an important impact on the stability of the foundation pit. Especially during the excavation process, the change of the underground water level may cause the soil to slip or become unstable. Therefore, installing underground water level sensors can monitor the change of the underground water level in real time and adjust the construction plan in time. The data collected by the sensors will be transmitted to the system to help judge the flow trend of the groundwater around the foundation pit.

[0017] Settlement monitoring sensors: Installed on the foundation pit and surrounding buildings to monitor the settlement of the foundation pit in real time. By collecting settlement data, settlement problems of the foundation pit or surrounding buildings can be detected early. The data of the settlement monitoring sensors will be used to correct the settlement amount in the simulation model and provide support for subsequent foundation pit stability analysis.

[0018] The collected data will be processed by the central processing unit to ensure the accuracy and effectiveness of the data. The data processing process includes: Data cleaning and preprocessing: The real-time transmitted data may contain outliers and noise. The system will perform preliminary cleaning and preprocessing on the data, including removing noise data, filling in missing data, and removing invalid data, etc.

[0019] Data normalization: In order to make the data from different sources consistent, all the data collected by the sensors will be normalized. This step can ensure that the values with different measurement units can be unified under the same dimension, providing a consistent data source for subsequent analysis.

[0020] Data storage and management: The data after cleaning and processing will be stored in the database. The database adopts a storage scheme combining relational databases and non-relational databases. Specifically, structured data (such as the specific values collected by sensors) is stored in the relational database, while unstructured data (such as pictures, reports, etc.) is stored in the non-relational database. This can ensure the efficient access and long-term stable storage of the data.

[0021] Data Synchronization and Real-time Monitoring: In some embodiments, to improve the timeliness and reliability of data transmission, the system transmits data to the cloud for synchronization processing and analysis in order to obtain information on the construction environment and foundation pit status in real time. This data synchronization technology can ensure the timeliness of monitoring data and simulation analysis results, and ensure real-time safety warning and decision support for foundation pit construction.

[0022] Multiple physical parameters and formula definitions involved in data collection will be used for modeling and analysis in subsequent steps. To ensure the accuracy of data and the establishment of subsequent simulation models, the following are the definitions of key parameters and their calculation formulas.

[0023] Elastic Modulus of Soil (E): The elastic modulus is used to describe the elastic deformation ability of soil after being stressed. In this step, the elastic modulus (E) is one of the key parameters obtained through geological exploration equipment, and the formula is as follows: Where: E is the elastic modulus of the soil (unit: Pa); σ is the stress of the soil (unit: Pa); ∈ is the strain of the soil (unit: dimensionless).

[0024] This formula describes the deformation characteristics of the soil in the elastic stage. By understanding the elastic modulus, it can help analyze the deformation and stress response of the soil during the foundation pit excavation process.

[0025] Groundwater Flow (Darcy's Law): The simulation of groundwater flow is an important part in the safety assessment of foundation pits. The basic formula for groundwater flow is Darcy's Law: Where: q is the water flow rate (unit: m 3 / s); K is the permeability coefficient of the soil (unit: m / s); A is the cross-sectional area of flow (unit: m 2 ); Δh is the hydraulic head difference (unit: m); L is the length of the water flow path (unit: m).

[0026] The groundwater flow formula can help the system accurately predict the impact of groundwater on the stability of the foundation pit, especially the fluctuation of the groundwater level during the excavation process.

[0027] Settlement (δ): The calculation of settlement is an essential part in the safety assessment of foundation pits. The calculation formula for settlement (δ) is as follows: Where: δ is the settlement (unit: m); P is the applied load (unit: N); L1 is the length of the load action (unit: m); A1 is the stressed area (unit: m 2 ); E is the elastic modulus of the soil (unit: Pa).

[0028] This formula is particularly important for settlement monitoring during the construction stage and can help to judge the settlement amount of the foundation pit and surrounding structures in real time.

[0029] In some embodiments, combined with the technology of wireless sensor network (WSN), the flexibility and scope of data collection will be further improved. By deploying multiple wireless sensor nodes in the foundation pit area, data can be transmitted in real time, and the stable transmission of data can be ensured through network protocols. The deployment range of sensors can be extended to the surrounding areas to comprehensively monitor the changes in the entire construction environment.

[0030] In addition, in order to better realize the real-time processing and analysis of data, the use of a cloud computing platform can ensure the rapid upload, storage, and analysis of data. After the data stream is transmitted from the on-site sensors to the cloud platform, it can be centrally processed through the cloud computing platform, which not only improves the data analysis efficiency but also ensures the security and reliability of the data.

[0031] By accurately collecting information such as the environmental data, soil properties, and groundwater changes of the foundation pit, a solid foundation is provided for subsequent modeling, analysis, and optimization.

[0032] S2. Establish a multi-physical-field coupling model of the foundation pit: According to the foundation pit engineering data collected in step S1, construct a multi-physical-field coupling model of the foundation pit. The model includes the interaction of soil mechanics, groundwater flow, and heat conduction, forming a coupled equation set for the stability analysis of the foundation pit; In step S2, the goal is to integrate these collected multi-dimensional data into a multi-physical-field coupling model to comprehensively describe the mechanical behavior, groundwater flow, thermal effects of the foundation pit, and their interactions. This multi-physical-field model can not only accurately simulate the behavior of the foundation pit at each stage during the construction process but also improve the accuracy and real-time performance of the model through effective coupling methods.

[0033] In this embodiment, by establishing a multi-physical-field coupling model, the model includes three major physical fields: soil mechanics, groundwater flow, and heat conduction, and considers their interactions. Each physical field has its specific mathematical model, and these models are linked through a coupled equation set to describe their dynamic responses during the actual construction process.

[0034] Soil mechanics model: To describe the mechanical behavior of soil under different loads and environmental conditions, a non-linear constitutive model is adopted, such as the Mohr-Coulomb model or the Hardening-Soil model. These models can accurately simulate the elastic and plastic deformations of soil, as well as the yield, hardening, and failure characteristics of soil. By establishing these models, the deformation, stress distribution, and possible failure modes of the soil during the excavation of the foundation pit can be predicted.

[0035] Groundwater flow model: The influence of groundwater on the foundation pit is particularly important. Especially during the excavation process, the change of the groundwater level may cause the mechanical properties of the soil mass to change. Groundwater flow is usually described by Darcy's law, and its core is that the hydraulic head difference drives the groundwater flow. By establishing a groundwater flow model around the foundation pit, the influence of water level change and water flow on the stability of the foundation pit can be simulated.

[0036] Heat conduction model: Although the thermal effect usually has a small influence on the mechanical properties of the soil mass, in special cases (such as extreme weather or large changes in groundwater temperature), the thermal effect may significantly change the physical properties of the soil mass. Therefore, this model incorporates the influence of temperature change into the analysis to ensure the comprehensiveness of the foundation pit stability analysis.

[0037] These physical field models are linked by a coupled equation set to form a complete simulation framework.

[0038] In order to combine soil mechanics, groundwater flow and heat conduction, this embodiment adopts the Lagrange multiplier method to constrain the interaction between different physical fields. The Lagrange multiplier method couples each physical field by introducing an additional constraint variable, so that the equations of these physical fields can be solved simultaneously under a unified framework.

[0039] Generally, the coupled equations will include the following types of equations: Soil mechanics equation: The mechanical behavior of the soil mass describes the deformation and stress response of the soil mass in the foundation pit under the action of external loads. The soil mechanics equation usually adopts the motion equation, similar to the classical Galerkin method, and its form is: Where: M eff is the effective mass matrix (unit: kg), representing the mass distribution of the soil mass under dynamic loads; is the displacement acceleration (unit: m / s 2 ), describing the acceleration of the soil mass under external loads; C is the damping matrix (unit: N·s / m), representing the internal friction and energy dissipation characteristics of the soil mass; is the displacement velocity (unit: m / s), describing the velocity of the soil mass under the action of loads; K eff is the effective stiffness matrix (unit: N / m), representing the elastic response of the soil mass. u is the displacement vector (unit: m), representing the displacement of the soil mass at different time steps; F ext is the external load (unit: N), such as the weight of the building, the load of construction equipment, etc.; F int Internal reaction force (unit: N), describing the distribution of internal forces in the soil mass.

[0040] Groundwater flow equation: The groundwater flow equation is based on Darcy's law and describes the basic law of groundwater flow in soil mass. It uses the "Darcy's law" disclosed in S1.

[0041] This equation can simulate the flow of water in the foundation pit area, especially the influence of groundwater on the stability of the foundation pit during the excavation process.

[0042] Heat conduction equation: The influence of heat conduction in soil mass will cause temperature changes in the soil mass during the excavation process, thus affecting the physical properties of the soil mass. The heat conduction equation is: Where: T is the temperature field (unit: °C or K), describing the temperature distribution of the soil mass; t is the time (unit: s), representing the process of heat conduction changing with time; k1 is the thermal conductivity (unit: W / (m·K)), describing the ability of the soil mass to conduct heat; is the temperature gradient (unit: K / m), describing the rate of change of temperature in space.

[0043] This equation is used to simulate the propagation of temperature in the soil mass, considering the potential influence of temperature on the stability of the soil mass.

[0044] Due to the complexity of the coupling between soil mechanics, groundwater flow and heat conduction, it may be very difficult to directly solve these coupled equations. Therefore, in this embodiment, the finite element method (FEM) is adopted to numerically solve these equations. In the finite element method, the entire foundation pit area is discretized into a finite number of small elements, and by solving the equations of each small element, the stress, displacement, settlement, temperature, water flow and other results of the entire foundation pit are finally obtained.

[0045] In order to improve the calculation efficiency and ensure the accuracy of the calculation results, this embodiment adopts the adaptive mesh optimization technology. According to the error evaluation in the simulation process, the mesh of the area with large stress changes is automatically refined, and the mesh of other areas is coarsened, so as to reduce the calculation amount and improve the calculation accuracy.

[0046] In some embodiments, in order to further improve the adaptability of the model, a heterogeneous permeability model can also be introduced to consider the differences in the permeability characteristics of different regions of the soil mass. Specifically, in different regions of the soil mass, the permeability coefficient K may have significant differences, and this difference has an important impact on groundwater flow. Therefore, considering the heterogeneous permeability of the soil mass in different regions during modeling will improve the simulation accuracy of groundwater flow.

[0047] In addition, a dynamic heat source model can also be considered to be introduced into the heat conduction equation. Especially under extreme weather conditions, the temperature change of the soil mass may affect the stability of the foundation pit. By dynamically adjusting the parameters of the heat source model, the accuracy of heat conduction analysis can be further improved.

[0048] By using the Lagrange multiplier method to constrain the interaction between these physical fields, the integrity and consistency of the coupled equations are ensured.

[0049] S3. Perform finite element analysis based on the multi-physical field coupling model: Based on the multi-physical field coupling model constructed in step S2, use the finite element method to simulate the mechanical behavior, groundwater flow, and thermal effects of the foundation pit, and obtain the analysis results of stress, displacement, and settlement. The core task of S3 is to perform detailed simulation analysis based on these coupling models using the finite element method, so as to accurately calculate the influence of the mechanical behavior, groundwater flow, and thermal effects of the foundation pit on the stability of the foundation pit. The ultimate goal of this step is to obtain key parameters such as the stress, displacement, and settlement of the foundation pit and evaluate the stability of the foundation pit.

[0050] First of all, by discretizing the foundation pit area into finite elements and solving the mechanical behavior, groundwater flow, and thermal effects of the foundation pit within these elements, the finite element method can effectively handle problems such as complex geometric shapes, material nonlinearities, and boundary conditions.

[0051] The finite element method (FEM), as a numerical calculation method, can transform complex continuum problems into discretized numerical problems and has wide applications in the fields of computational structural mechanics, fluid mechanics, and heat conduction. In the simulation analysis of the foundation pit, the implementation steps of the finite element method include: Discretization process: In finite element analysis, the foundation pit area first needs to be discretized into multiple small elements. The shapes of these small elements are usually triangles or quadrilaterals (for two-dimensional problems), or tetrahedrons or hexahedrons (for three-dimensional problems). Physical quantities (such as stress, displacement, temperature, etc.) will be calculated within each small element. Through this discretization method, the entire area of the foundation pit is converted into a grid composed of multiple elements.

[0052] Boundary conditions and loading conditions: To simulate the behavior of the foundation pit in the real environment, the boundary conditions and loading conditions of the model must be accurately set. Boundary conditions usually include the supports around the foundation pit, the constraints on the excavation surface, etc. Loading conditions include the self-weight in the foundation pit, construction loads, and changes in the external environment, etc.

[0053] In some embodiments, the load condition may be a dynamic load, that is, the load changes with time. For example, during the excavation of the foundation pit, the construction equipment load and the self-weight of the upper soil layer may change with the progress of construction, and the application of the dynamic load model can better reflect this change.

[0054] Solve the finite element equation: The finite element analysis of the foundation pit is numerically solved based on the following basic equation: [K]·{u}={F ext +F fluid}; Where: [K] is the stiffness matrix (unit: N / m), representing the response of the system to external loads; {u} is the displacement vector (unit: m), describing the displacements of each node in the foundation pit; {F ext} is the external load (unit: N); {F fluid} is the additional force caused by water flow (unit: N), which is the force generated by the water pressure caused by groundwater flow.

[0055] Coupled equation solution: In step S2, how to consider the interaction between soil mechanics, groundwater flow, and heat conduction through the coupled equation has been introduced. When solving by the finite element method, the coupling effect of each physical field during the solution process is solved simultaneously through the relevant matrix equations.

[0056] During the foundation pit analysis process, the interaction between multiple physical fields (such as soil mechanics, groundwater flow, and heat conduction) cannot be ignored. Therefore, during the simulation of the finite element method, the coupling effect between these physical fields must be considered simultaneously.

[0057] Coupling of soil mechanics and groundwater flow: Groundwater flow will generate water pressure on the soil, thereby affecting the deformation and stress distribution of the soil. This coupling can be described by the following formula: F fluid = ∫ A Δh·K·ndA; Where: F fluid is the additional force caused by water flow (unit: N), the pore pressure generated by groundwater flow; Δh is the head difference (unit: m), the pressure difference between two points of water flow; K is the permeability coefficient of the soil (unit: m / s); n is the normal vector (unit: dimensionless), indicating the direction of the force action; A is the flow cross-sectional area (unit: m 2 ); dA is the area element (unit: m 2 ).

[0058] Coupling of soil mechanics and heat conduction: The temperature change of the soil will affect its elastic modulus and yield strength, thereby changing its mechanical response. The additional force caused by the thermal effect can be expressed by the following formula: F thermal = ∫ A α·ΔT1·n1dA2; Where: F thermal is the additional force caused by the thermal effect (unit: N), the force generated by thermal expansion caused by temperature change. α is the coefficient of thermal expansion (unit: 1 / K), describing the thermal expansion characteristics of the soil; ΔT1 is the temperature change (unit: °C or K), representing the temperature change of the soil during excavation due to the thermal effect; n1 is the normal vector (unit: dimensionless), indicating the direction of deformation caused by thermal expansion; A2 is the area of the region (unit: m2 ) to describe the cross-sectional area affected by thermal expansion.

[0059] To accelerate the finite element solution process, especially in the simulation of large-scale foundation pit projects, parallel computing technology is adopted. By distributing the computing tasks to multiple processing units (such as multi-core CPUs or cloud computing platforms), multiple sub-problems can be processed simultaneously, significantly improving the computing efficiency.

[0060] In some embodiments, adaptive mesh optimization is introduced into the calculation process. This technology automatically adjusts the refinement degree of the mesh according to the stress changes and physical properties within the foundation pit area. In areas with stress concentration, the mesh will be refined to improve the computing accuracy; in relatively stable areas, the mesh will remain coarser to reduce the computational amount.

[0061] In some embodiments, as the construction environment changes, it may be necessary to adjust the simulation model in real time. Based on real-time monitoring data (such as groundwater level, settlement, temperature, etc.), the relevant parameters in the finite element model can be dynamically updated to ensure the accuracy of the simulation analysis.

[0062] In addition, the introduction of a heterogeneous soil layer model can further improve the accuracy of groundwater flow and heat conduction analysis. In practical applications, the physical properties (such as permeability, thermal conductivity) of different soil layers may vary significantly, and these differences have an important impact on the stability of the foundation pit. Therefore, considering the heterogeneous characteristics of different soil layers is crucial for optimizing the simulation model.

[0063] By optimizing the calculation process, introducing parallel computing technology and adaptive mesh optimization methods, the computing efficiency and accuracy are further improved, providing a scientific basis for the safety assessment and construction optimization of foundation pit projects.

[0064] S4. Conduct nonlinear finite element analysis and adaptive mesh optimization: According to the finite element analysis results in step S3, further use a nonlinear constitutive model to accurately model the soil mass, and use an error estimation method to adaptively optimize the mesh; In the foregoing technical solutions, the finite element method is used to solve the multi-physical field coupling model of the foundation pit, and the stress, displacement, settlement, etc. of the foundation pit at different construction stages are obtained. However, the behavior of the foundation pit soil mass usually exhibits nonlinear characteristics. Especially during the excavation process, when the soil mass bears a large load, the nonlinear effects that occur will have an important impact on the stability of the foundation pit. Therefore, in S4, the key task is to conduct more refined modeling of the soil mass through nonlinear finite element analysis, and at the same time adopt an adaptive mesh optimization method to ensure high computing accuracy in key areas and improve the overall computing efficiency.

[0065] In this embodiment, first, a non-linear constitutive model, especially the Mohr-Coulomb or Hardening-Soil model, is adopted to accurately describe the non-linear behavior of the soil mass. Then, based on the aforementioned model and analysis results, an error estimation method is used for adaptive mesh optimization. In this way, the system can dynamically adjust the mesh density according to the stress and displacement distributions, thereby reducing the waste of computing resources while ensuring the calculation accuracy.

[0066] To accurately describe the non-linear behavior of the soil mass under different loads, this embodiment adopts the commonly used Mohr-Coulomb model or Hardening-Soil model. These models can effectively capture the non-linear characteristics of the soil mass such as yielding, hardening, and softening. Especially during the foundation pit excavation process, the stress on the soil mass often exceeds the yield point, resulting in plastic flow or hardening of the soil mass. Therefore, adopting a non-linear constitutive model helps to improve the accuracy of foundation pit simulation analysis.

[0067] The constitutive relationship of the Mohr-Coulomb model is as follows: σ = σ yield + k2·∈; Where: σ is the stress (unit: Pa), representing the total stress in the soil mass; σ yield is the yield stress (unit: Pa), which is the stress at which the soil mass begins plastic deformation; k2 is the hardening coefficient (unit: Pa), used to describe the change in stiffness during the plastic deformation of the soil mass; ∈ is the strain (unit: dimensionless), describing the degree of deformation of the soil mass.

[0068] By modeling the stress-strain relationship of the soil mass, this model can effectively simulate the possible plastic deformation during the foundation pit excavation process.

[0069] The solution process in non-linear finite element analysis usually adopts the Newton-Raphson method. Since the stiffness matrix in the non-linear equation system changes with the displacement, the Newton-Raphson method can iteratively solve the non-linear equation to gradually obtain the true values of stress and displacement.

[0070] In each iteration, the system updates the displacement through the following formula: K(u)·Δu = F - F current ; Where: K(u) is the non-linear stiffness matrix (unit: N / m), which depends on the current displacement field and represents the system's response to the load; Δu is the displacement increment (unit: m), representing the amount of displacement change in each iteration; F is the external load vector (unit: N), representing the external load applied to the foundation pit; F current$\mathbf{F}$ is the current load vector (unit: N), representing the load calculated from the current displacement.

[0071] By iterative solution until the displacement increment $\Delta\mathbf{u}$ converges, the final displacement, stress distribution, and other results of the foundation pit can be obtained.

[0072] In complex nonlinear analysis, especially in stress concentration areas, the choice of mesh density is crucial. Traditional uniform mesh division may not be able to fully capture the stress changes in these areas. Therefore, in this embodiment, an adaptive mesh optimization method is adopted to dynamically adjust the mesh resolution according to the calculation results.

[0073] The key steps of adaptive mesh optimization include: Error estimation: By evaluating the errors of stress, displacement, or other physical quantities of each element, it is determined which areas need to refine the mesh.

[0074] Mesh refinement and coarsening: According to the error evaluation results, the mesh is refined in the stress concentration areas, while the mesh is coarsened in the stable areas.

[0075] The formula for error estimation is: where: $\eta$ is the error estimation (unit: dimensionless), representing the error ratio in a certain calculation area; $\mathbf{u}$ error is the error of stress or displacement (unit: m or Pa), the error amount in the calculation result; $\mathbf{u}$ total is the total stress or displacement of the calculation result (unit: m or Pa), describing the stress or displacement of the soil mass.

[0076] This formula can help the system evaluate the calculation accuracy of different areas, so as to determine the refinement and coarsening of the mesh.

[0077] The computational amount of nonlinear finite element analysis and adaptive mesh optimization is usually very large, especially in large-scale foundation pit projects. To improve the calculation efficiency, this embodiment adopts parallel computing technology to accelerate the calculation process by distributing the calculation tasks to multiple computing nodes.

[0078] In some embodiments, optimization is carried out through a multi-level parallel strategy. Specifically, the calculation tasks can be parallel not only at the node level but also at the data level for block processing, and each computing node is responsible for processing a part of the data. This strategy can effectively improve the calculation efficiency and reduce the overall calculation time.

[0079] In addition to the basic nonlinear analysis and mesh optimization methods, further optimization can also be carried out under different conditions. Considering the different material properties between soil layers is crucial for the accuracy of the model.

[0080] In addition, to cope with dynamic loads, such as changes in construction equipment loads, the system can update the load model in real time by combining real-time sensor data, making the simulation results closer to the actual construction situation.

[0081] By applying the non-linear constitutive model, the complex deformation behavior of the soil under load can be accurately described. And through adaptive mesh optimization, while improving the calculation accuracy, unnecessary consumption of computing resources can be reduced. In addition, combined with parallel computing technology, the challenges of large-scale computing tasks are effectively solved, ensuring that the safety assessment of the foundation pit can be carried out efficiently and accurately.

[0082] S5. Optimize the parameters of the simulation model and perform deep learning optimization: Based on the finite element analysis results of step S4, use a convolutional neural network to learn the historical data and real-time monitoring data, and automatically optimize the parameter settings of the simulation model to enhance the simulation accuracy. In the foregoing steps, through non-linear finite element analysis and adaptive mesh optimization, a detailed simulation analysis of the stability of the foundation pit has been carried out. However, as the complexity of the model increases, it becomes increasingly difficult to manually adjust the model parameters, and traditional calculation methods may require a large amount of computing resources and time. Therefore, in S5, it is proposed to automate the parameter adjustment and improve the simulation accuracy by introducing deep learning optimization.

[0083] In this embodiment, based on the finite element analysis results obtained in step S4, a convolutional neural network (CNN) is used to learn the historical data and real-time monitoring data. By training the neural network, the system can automatically optimize the parameter settings in the simulation model, improve the accuracy of the simulation analysis, and make the model more efficient in dealing with complex and changing actual situations.

[0084] Deep learning models, especially convolutional neural networks (CNNs), perform excellently in image recognition and complex pattern recognition. For the foundation pit simulation analysis, CNN is used to extract features from a large amount of historical case data and real-time monitoring data, and automatically adjust the parameters of the model. This not only reduces manual intervention but also improves the prediction ability and adaptive ability of the model.

[0085] In some embodiments, the system optimizes the following model parameters through deep learning training using historical data and real-time data: The constitutive model parameters of the soil (such as hardening coefficient, yield stress, etc.); The permeability coefficient and the head difference, which affect the groundwater flow model; The coefficient of thermal expansion, which affects the heat conduction model; The loads and boundary conditions to simulate the changes in the environment and external conditions during the construction process.

[0086] Through the optimization of the deep learning model, these parameters can be automatically adjusted to ensure that the calculation results of the simulation model are more accurate at each stage, thereby providing a scientific basis for engineering decisions in different construction stages.

[0087] By adopting a Convolutional Neural Network (CNN), this network structure is particularly suitable for processing input data with spatial structures, such as images, sensor data, etc. Through multiple convolutional layers, pooling layers, and fully connected layers, CNN can extract feature information in the data layer by layer. In the foundation pit simulation, CNN can learn the relationships between the mechanical responses of soil, groundwater flow, and thermal effects, etc. from historical case data, and optimize the model based on real-time monitoring data.

[0088] Input data: Historical data includes the construction plan of the foundation pit, soil layer distribution, environmental monitoring data (such as climate, groundwater level, etc.), and the corresponding construction results (such as settlement, stress, etc.). Real-time monitoring data includes information such as stress, displacement, and temperature collected during the construction process.

[0089] Feature extraction and learning: CNN automatically extracts spatial features in the input data through multiple convolutional layers, such as the stress distribution of soil and the flow state of groundwater. Through the pooling operation, the network can gradually compress the data and retain important feature information.

[0090] Parameter optimization: Through the backpropagation algorithm, the network adjusts the weights according to the error in the training data to optimize the parameters in the model. Finally, the trained model can automatically adjust the parameter settings in the simulation model according to new input data (such as construction progress and environmental changes).

[0091] The backpropagation algorithm is a commonly used optimization algorithm in deep learning, and its basic process is as follows: where: w new is the optimized network weight; w old is the current network weight; η is the learning rate (unit: dimensionless), which controls the step size of each weight update; is the gradient of the loss function (unit: dimensionless), indicating the contribution of the current weight to the error.

[0092] Through the backpropagation algorithm, CNN can gradually adjust its parameters to make the simulation results more in line with the actual situation and improve the prediction accuracy.

[0093] During the training process, the system continuously optimizes the model parameters by monitoring real-time feedback data (such as settlement, groundwater level changes, climate changes, etc.). This adaptive optimization process can automatically adjust the non-linear parameters of the model to ensure that the model still maintains high accuracy in the face of different construction conditions. In this way, the system can update and adapt to any changes in the foundation pit construction process in real time, avoiding the cumbersome process of manual parameter adjustment.

[0094] In some embodiments, the deep learning model not only automatically adjusts the physical model parameters, but also can verify and optimize the simulation results under different construction scenarios. For example: Groundwater flow model: Optimize parameters such as head difference and permeability coefficient through the deep learning model to better simulate the flow changes of groundwater.

[0095] Soil constitutive model: Automatically adjust parameters such as hardening coefficient and yield stress to ensure that the non-linear behavior of the soil is accurately described.

[0096] Thermal effect model: Automatically adjust the thermal expansion coefficient and the influence of temperature changes on the soil according to real-time climate data and groundwater level changes.

[0097] In practical applications, this embodiment can also introduce deep reinforcement learning, enabling the model to continuously self-adjust and optimize according to the real-time changing environmental conditions during the construction process.

[0098] In addition, as the complexity of the foundation pit project increases, the training data of the deep learning model can be continuously expanded and updated. By continuously incorporating new engineering cases and monitoring data, the deep learning model can be continuously optimized during the construction process, improving the accuracy of its assessment of the foundation pit stability.

[0099] Through the learning and optimization of historical data and real-time monitoring data by the convolutional neural network (CNN), the simulation model can be adjusted in real time during the construction process to ensure the accuracy and reliability of the foundation pit safety assessment.

[0100] S6. Perform cloud computing parallel computing optimization: Based on the optimized model obtained in step S5, perform parallel computing through the cloud computing platform, and distribute the simulation tasks to multiple computing nodes to ensure the efficient processing of large-scale computing tasks; In the above steps, the parameters of the simulation model are optimized based on deep learning, enabling the foundation pit simulation analysis to automatically adjust and improve the simulation accuracy under different environmental conditions. However, due to the scale and complexity of foundation pit engineering, especially in large-scale foundation pit analysis, the computing power of a single computer may not meet the requirements of real-time analysis and efficient simulation calculation. Therefore, S6 introduces a cloud computing platform and parallel computing technology to solve this problem. By distributing computing tasks to multiple computing nodes, cloud computing parallel computing can significantly accelerate the simulation process, improve the analysis efficiency, and reduce the consumption of hardware resources at the same time.

[0101] To further improve the computing efficiency, the foundation pit simulation task is decomposed into multiple subtasks and executed simultaneously through the distributed computing ability of the cloud computing platform. In addition, the cloud computing platform can dynamically allocate computing resources to ensure the smooth progress of computing tasks under different load conditions.

[0102] In the cloud computing platform, the dynamic scheduling of elastic computing resources is the key to improving computing efficiency and reducing resource waste. According to different computing requirements and task loads, the cloud platform will adjust computing resources in real time. For parallel computing tasks, the optimization of task allocation is crucial for improving the overall computing efficiency.

[0103] The formula for elastic computing resource scheduling and load balancing is as follows: Where: R task is the computing resource allocated to each task (unit: dimensionless). According to the complexity and computing requirements of the task, the computing power that each subtask can obtain is determined; R available is the total available computing resource (unit: dimensionless). It represents the total computing power available for use by the cloud platform at the current moment, including processors, memory, storage, etc.; N is the number of computing nodes (unit: dimensionless). It represents the number of computing nodes used by the cloud computing platform during parallel computing.

[0104] This formula describes the reasonable allocation of computing resources. By evenly distributing the total available computing resources according to the number of tasks, each task can obtain appropriate computing resources, thus accelerating the overall simulation analysis process.

[0105] During parallel computing, the computing task is decomposed into multiple subtasks and processed in parallel through multiple computing nodes. These subtasks can be stress analysis in different regions, calculations at different time steps, etc. Through reasonable subtask allocation, the system can significantly improve the speed and efficiency of simulation calculation.

[0106] The task allocation formula is as follows: Where: T iis the subtask assigned to the i-th node (unit: dimensionless); T total is the total computing task volume (unit: dimensionless); M i is the node weight (unit: dimensionless).

[0107] This formula ensures that each computing node matches its computing capacity in terms of the amount of tasks assigned to it, thereby optimizing load balancing and computing efficiency during the computing process.

[0108] In data-level parallel computing, model data is divided into blocks, and each computing node is responsible for processing part of the data block. Through parallel computing, multiple computing nodes process different data blocks at the same time, improving the overall computing efficiency.

[0109] The data segmentation and processing formula is as follows: Where: D i The data block processed by the i-th computing node (unit: dimensionless); D total is the total data volume (unit: dimensionless); P i is the node processing capacity (unit: dimensionless).

[0110] This formula means that the total amount of data is reasonably distributed according to the processing capacity of each node, ensuring that each node is not overloaded when processing data, thereby optimizing the parallel computing process.

[0111] In order to ensure the load balance of each node during parallel computing, the cloud computing platform will dynamically adjust the task allocation to ensure that the load of each node is as equal as possible, thereby avoiding excessive load on some nodes affecting the overall computing efficiency.

[0112] The load balancing calculation formula is as follows: Where: L i is the load of the ith node (unit: dimensionless); T i is the amount of tasks assigned to the i-th node (unit: dimensionless); C i is the computing power of the ith node (unit: dimensionless).

[0113] This formula is used to calculate the load of each node to ensure that when allocating tasks, the load on a certain node is avoided to be too high or too low, thereby improving computing efficiency.

[0114] In order to cope with the dynamic changes in computing load, the cloud computing platform will dynamically adjust computing resources according to real-time computing needs to ensure that the system's computing power always matches the needs of the computing tasks.

[0115] The dynamic resource scheduling formula is as follows: R assigned = R available ·F demand ; Where: R assigned is the computing resource allocated to the current task (unit: dimensionless); R available is the currently available computing resource (unit: dimensionless); F demand is the demand factor of the current computing task (unit: dimensionless).

[0116] Through this formula, the cloud platform can dynamically allocate resources according to the requirements of computing tasks, ensuring that the resource allocation during the computing process will not cause bottlenecks due to fluctuations in computing load.

[0117] The parallel computing and load balancing strategy of this embodiment can be combined with real-time monitoring data to further optimize the resource scheduling and task allocation of the cloud computing platform.

[0118] In addition, as the scale of computing tasks increases, the multi-level parallel computing and deep learning optimization of the cloud computing platform can also be used in combination. By adjusting the parameters and computing strategies of the simulation model through the real-time feedback of the deep learning model, the efficiency and accuracy of the simulation analysis can be further improved.

[0119] With the support of dynamic allocation of computing resources and load balancing, the system can efficiently and stably process large-scale foundation pit simulation tasks, ensuring the efficiency and accuracy of engineering calculations. Through the feedback of real-time monitoring data, the system can automatically adjust computing resources and simulation models, providing accurate safety assessments and optimization suggestions for foundation pit construction.

[0120] S7. Output the foundation pit safety assessment report and provide optimization suggestions: Generate a foundation pit safety assessment report based on the simulation results obtained in step S6, and provide optimization suggestions according to the safety assessment results; In S7, the system will generate a foundation pit safety assessment report based on these simulation results and provide optimization suggestions according to the assessment results to ensure the safety of foundation pit construction.

[0121] The system automatically generates a safety assessment report for the foundation pit. The report content is based on the simulation results and details the stability of the foundation pit at each construction stage. Specifically, the report will analyze key indicators such as stress distribution, displacement changes, and settlement conditions at different construction stages, evaluate potential risk points, and propose countermeasures. In addition, the report will also propose corresponding optimization suggestions based on the actual situation of foundation pit construction, combined with historical data and real-time monitoring information.

[0122] In this embodiment, the generation of the foundation pit safety assessment report mainly depends on the following aspects: Stress analysis: When evaluating the stability of the foundation pit, the stress in each area of the foundation pit is first analyzed. By interpreting the stress data in the simulation results, it is evaluated whether the structure of the foundation pit can withstand the loads generated during the construction process.

[0123] Displacement and settlement analysis: In addition to stress analysis, the displacement and settlement of the foundation pit are important factors affecting stability. During the construction process, the deformation of the soil around the foundation pit may cause uneven settlement, which in turn affects the structural stability. The report will combine the displacement data in the simulation results to evaluate whether excessive deformation has occurred in the foundation pit and analyze the settlement risk based on the nonlinear behavior of the soil.

[0124] Analysis of the influence of groundwater flow: In some embodiments, the groundwater flow has an important impact on the stability of the foundation pit. The change of the groundwater level may cause changes in the strength and stiffness of the soil, which in turn affects the stability of the foundation pit. Therefore, the foundation pit safety assessment report also includes an analysis of the influence of groundwater flow on the stability of the foundation pit. The system will evaluate the additional forces that the water flow may bring and the impact on the foundation pit structure according to the simulation results of the groundwater level change and flow.

[0125] Thermal effect analysis: For some specific foundation pit construction environments, such as high-temperature or low-temperature areas, the thermal effect may also affect the soil, causing the soil to expand or contract, thereby triggering instability. The system generates a thermal effect analysis report by simulating and analyzing the deformation behavior of the soil at different temperatures.

[0126] The foundation pit safety assessment report not only provides the analysis results of the current foundation pit state, but also generates optimization suggestions according to the analysis results to help the construction party improve the design or take emergency measures.

[0127] Optimization suggestions for the support system: In some embodiments, the support system of the foundation pit may need to be further strengthened. The report will propose suggestions to increase the number and layout density of the support piles according to the results of the stress distribution and settlement analysis.

[0128] Soil reinforcement plan: For some foundation pits with relatively soft or uneven soil, soil reinforcement may be required. The report will propose a soil reinforcement plan according to the mechanical properties of the soil and the influence of groundwater flow, such as grouting, soil nailing wall, deep mixing and other technologies, to enhance the stability and bearing capacity of the soil.

[0129] Groundwater level control plan: For foundation pits greatly affected by groundwater flow, the report may propose a groundwater level control plan. By adjusting the drainage system or strengthening the waterproof measures of the retaining structure, the adverse impact of groundwater on the foundation pit can be avoided.

[0130] Monitoring and control measures: In addition to soil reinforcement and optimization of the support system, the report will also propose suggestions for real-time monitoring to keep track of the safety status during the foundation pit construction at any time.

[0131] In this embodiment, the generation of the foundation pit safety assessment report depends on the data processing module in the system. First, the system obtains relevant engineering data and construction information from the historical case database and calibrates the simulation results in combination with the real-time monitoring data. Subsequently, the system generates the report through the following steps: Data integration and analysis: Compare the simulation results with the actual monitoring data to ensure the accuracy of the simulation results. Based on the data analysis results, the system judges the stability of the foundation pit at different construction stages.

[0132] Risk assessment: According to the analysis results, the system automatically identifies possible risk points, such as excessive stress, rapid settlement, etc., and assesses the possible unstable situations of the foundation pit.

[0133] Generation of optimization suggestions: After identifying the risks, the system automatically generates targeted optimization suggestions based on historical cases and engineering experience, such as increasing supports, adjusting the groundwater level, etc.

[0134] Automatic report generation: Finally, the system integrates all the analysis results and optimization suggestions into a complete report. The report includes detailed charts and data, as well as corresponding text descriptions. The report format is standardized for easy reading and reference by users.

[0135] S7 generates a foundation pit safety assessment report based on the simulation analysis results and provides optimization suggestions in combination with real-time data, providing a detailed safety assessment and improvement measures for the construction unit. This report not only provides an analysis of the current state of the foundation pit, but also proposes effective solutions for potential risks, helping to ensure the safety of foundation pit construction and the engineering quality.

[0136] The foundation pit simulation analysis system based on the finite element method described below can be mutually referred to the foundation pit simulation analysis method based on the finite element method described above.

[0137] Please refer to the appendix Figure 2 , the present invention also provides a foundation pit simulation analysis system based on the finite element method, including: Data acquisition module: This module uses geological exploration equipment, environmental monitoring sensors, climate monitoring equipment, groundwater level sensors and settlement monitoring sensors to collect and store the real-time data of the foundation pit project; Multi-physical field coupling modeling module: According to the collected foundation pit project data, establish a multi-physical field coupling model including soil mechanics, groundwater flow and heat conduction; Finite Element Analysis Module: Based on the multi-physics field coupling model, the finite element method is used to simulate and analyze the mechanical behavior, groundwater flow, and thermal effects of the foundation pit; Nonlinear Analysis and Mesh Optimization Module: The Mohr-Coulomb nonlinear constitutive model is used for nonlinear analysis of soil, and the finite element mesh is optimized according to error estimation. This module ensures fine calculation in areas with stress concentration by automatically adjusting the mesh density, while coarsening the mesh in other areas; Deep Learning Optimization Module: Convolutional neural networks are used to learn historical data and real-time monitoring data, and automatically optimize the parameters of the simulation model; Cloud Computing and Parallel Computing Module: Based on the optimized model, a cloud computing platform is used for distributed parallel computing. It adopts elastic computing resources, dynamically allocates tasks according to the computing load, and uses load balancing technology to ensure uniform load on each computing node and avoid single-point overload; Report Generation and Optimization Suggestion Module: A safety assessment report for the foundation pit is generated based on the simulation results. The report content includes the stability analysis of the foundation pit, construction risk assessment, and safety assessment of each construction stage.

[0138] The system of this embodiment can be used to execute the above method embodiment, and its principle and technical effects are similar, so they will not be elaborated here.

[0139] The foundation pit simulation analysis device based on the finite element method described below can be correspondingly referred to the foundation pit simulation analysis method based on the finite element method described above.

[0140] Please refer to the appendix Figure 3 , the present invention also provides a foundation pit simulation analysis device based on the finite element method, including: Data Acquisition Equipment: Includes geological exploration equipment, environmental monitoring sensors, groundwater level sensors, and settlement monitoring sensors; Computing and Simulation Processing Equipment: Adopts high-performance computer hardware, including a processor, memory, and hard disk storage, and supports efficient processing of parallel computing tasks; Cloud Computing Platform: A cloud platform that provides elastic computing resources and supports cloud storage and computing; Deep Learning Processing Unit: Equipped with a deep learning processing unit for training convolutional neural networks and optimizing the parameters of foundation pit simulation; Output and Feedback Equipment: Includes a display screen, printing equipment, and network interface for displaying simulation results and generating a safety assessment report for the foundation pit.

[0141] The device of this embodiment can be used to execute the above method embodiment, and its principle and technical effects are similar, so they will not be elaborated here.

[0142] Please refer to the appendix Figure 4, the present invention also provides a computer-readable storage medium, on which program code is stored. When the program code is executed, it causes a computer to execute a foundation pit simulation analysis method based on the finite element method. The storage medium includes: Data collection and storage instructions for controlling data acquisition devices to collect and store real-time data of foundation pit projects; Multi-physical field coupling model construction instructions for constructing a multi-physical field coupling model of the foundation pit based on the collected foundation pit project data and generating a coupling equation set; Finite element simulation analysis instructions for performing finite element method simulation calculations on the basis of the constructed multi-physical field coupling model to obtain analysis results of stress, displacement, and settlement; Nonlinear finite element analysis and mesh optimization instructions for accurately modeling the soil body using a nonlinear constitutive model and performing adaptive mesh optimization according to the finite element simulation results; Deep learning optimization instructions for using a convolutional neural network to learn historical data and real-time monitoring data and automatically optimizing the parameter settings of the simulation model; Cloud computing and parallel computing optimization instructions for performing parallel computing through a cloud computing platform, distributing simulation tasks to multiple computing nodes, and ensuring the rapid processing of large-scale computing tasks.

[0143] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The foundation pit simulation analysis method based on the finite element method is characterized by: The following steps are involved: S1. Collect foundation pit engineering data and environmental information: Use geological exploration equipment, environmental monitoring sensors, climate monitoring equipment, groundwater level sensors and settlement monitoring sensors to collect and store real-time monitoring data on the geological conditions, soil mechanical properties, groundwater level, climate change and settlement of foundation pit engineering; S2. Establishing a multi-physics field coupling model of the foundation pit: constructing a multi-physics field coupling model of the foundation pit according to the foundation pit engineering data collected in step S1, wherein the model includes the interaction of soil mechanics, groundwater flow and heat conduction to form a coupling equation group for foundation pit stability analysis; S3. Perform finite element analysis based on the multi-physics coupling model: Based on the multi-physics coupling model constructed in step S2, the finite element method is used to simulate the mechanical behavior, groundwater flow and thermal effect of the foundation pit to obtain stress, displacement and settlement analysis results; S4, performing nonlinear finite element analysis and adaptive mesh optimization: according to the finite element analysis results in step S3, further using a nonlinear constitutive model to accurately model the soil, and using an error estimation method to adaptively optimize the mesh; S5. Optimize simulation model parameters and perform deep learning optimization: Based on the finite element analysis results described in step S4, use a convolutional neural network to learn historical data and real-time monitoring data, and automatically optimize the parameter settings of the simulation model to enhance the simulation accuracy; S6, perform cloud computing parallel computing optimization: Based on the optimization model obtained in step S5, perform parallel computing through the cloud computing platform, distribute the simulation tasks to multiple computing nodes, and ensure efficient processing of large-scale computing tasks; S7. Output foundation pit safety assessment report and provide optimization suggestions: Generate a foundation pit safety assessment report based on the simulation results obtained in step S6, and provide optimization suggestions based on the safety assessment results.

2. The foundation pit simulation analysis method based on the finite element method according to claim 1 is characterized in that: The multi-physics coupling model includes: Coupled equations of soil mechanics and groundwater flow; Coupled equations of soil mechanics and heat conduction; The solution method based on the Lagrange multiplier method ensures the dynamic coupling between mechanics, fluid and heat conduction.

3. The foundation pit simulation analysis method based on finite element method according to claim 1 is characterized in that: The nonlinear finite element analysis and adaptive mesh optimization include: The nonlinear constitutive model is used to describe the nonlinear behavior of soil, and the relationship between foundation pit load and deformation is analyzed; A mesh optimization method based on error estimation refines the mesh in areas with severe deformation and coarsens the mesh in other areas; Dynamic mesh refinement and coarsening adjusts mesh resolution by automatically detecting stress changes and geological features to improve computational accuracy and efficiency.

4. The foundation pit simulation analysis method based on the finite element method according to claim 1 is characterized in that: The deep learning optimization simulation model includes: The convolutional neural network model is used to extract features from foundation pit soil data, groundwater level, and climate change input data, and the simulation parameters are automatically optimized through the training process; The weights in the convolutional neural network model are adjusted through the back-propagation algorithm to reduce the error of the simulation results and automatically adjust the model parameters according to real-time data.

5. The foundation pit simulation analysis method based on finite element method according to claim 1 is characterized in that: The cloud computing and parallel computing optimization include: Divide the foundation pit simulation analysis task into multiple subtasks and perform distributed computing through the cloud computing platform to speed up the computing speed; Utilize elastic computing resources to dynamically allocate computing tasks according to computing load and improve computing efficiency; Use load balancing technology to ensure that each computing node has a uniform load and avoid single-point overload.

6. The foundation pit simulation analysis method based on finite element method according to claim 1 is characterized in that: The foundation pit safety assessment report includes: Evaluate the stability of the foundation pit at each construction stage based on the simulation results; Generate construction optimization suggestions based on safety assessment results, including additional support piles or soil reinforcement solutions; Provide risk warning information and provide real-time response measures to changes in the construction environment.

7. The foundation pit simulation analysis method based on finite element method according to claim 1 is characterized in that: The historical data and real-time monitoring data include: Soil settlement, groundwater flow level, vibration, and temperature sensor data during construction; Geological data, engineering design parameters, and construction monitoring results of historical foundation pit cases; Use a combination of historical data and real-time monitoring data to improve and optimize the simulation analysis process.

8. The foundation pit simulation analysis system based on the finite element method is characterized by: The foundation pit simulation analysis method based on the finite element method according to any one of claims 1 to 7 comprises: Data acquisition module: This module uses geological exploration equipment, environmental monitoring sensors, climate monitoring equipment, groundwater level sensors, and settlement monitoring sensors to collect and store real-time data of foundation pit engineering; Multi-physics coupling modeling module: Based on the collected foundation pit engineering data, a multi-physics coupling model including soil mechanics, groundwater flow and heat conduction is established; Finite element analysis module: Based on the multi-physics field coupling model, the finite element method is used to simulate and analyze the mechanical behavior, groundwater flow and thermal effects of the foundation pit; Nonlinear analysis and mesh optimization module: Use the Mohr-Coulomb nonlinear constitutive model to perform nonlinear analysis of soil and optimize the finite element mesh based on error estimation. This module automatically adjusts the mesh density to ensure fine calculation in stress concentration areas and coarsens the mesh in other areas. Deep learning optimization module: Use convolutional neural networks to learn historical data and real-time monitoring data, and automatically optimize the parameters of the simulation model; Cloud computing and parallel computing module: Based on the optimization model, the cloud computing platform is used for distributed parallel computing. It uses elastic computing resources, dynamically allocates tasks according to computing load, and uses load balancing technology to ensure that the load of each computing node is even, avoiding single point overload; Report generation and optimization suggestion module: Generate a foundation pit safety assessment report based on the simulation results. The report content includes foundation pit stability analysis, construction risk assessment, and safety assessment of each construction stage.

9. The foundation pit simulation analysis equipment based on the finite element method is characterized by: The foundation pit simulation analysis method based on the finite element method according to any one of claims 1 to 7 comprises: Data acquisition equipment: including geological exploration equipment, environmental monitoring sensors, groundwater level sensors and settlement monitoring sensors; Computing and simulation processing equipment: using high-performance computer hardware, including processors, memory and hard disk storage, to support efficient processing of parallel computing tasks; Cloud computing platform: a cloud platform that provides elastic computing resources and supports cloud storage and computing; Deep learning processing unit: equipped with a deep learning processing unit for training convolutional neural networks and optimizing foundation pit simulation parameters; Output and feedback equipment: including display screen, printing equipment and network interface, used to display simulation results and generate foundation pit safety assessment report.

10. A computer-readable storage medium, characterized in that: A program code is stored thereon, and when the program code is executed, the computer is caused to execute a foundation pit simulation analysis method based on the finite element method, and the storage medium includes: Data collection and storage instructions, used to control data acquisition equipment to collect and store real-time data of foundation pit engineering; A multi-physics coupling model building instruction is used to build a multi-physics coupling model of the foundation pit based on the collected foundation pit engineering data and generate a coupling equation group; Finite element simulation analysis command, used to perform finite element simulation calculations based on the constructed multi-physics field coupling model to obtain analysis results of stress, displacement and settlement; Nonlinear finite element analysis and mesh optimization commands are used to accurately model the soil using nonlinear constitutive models based on finite element simulation results and perform adaptive mesh optimization; Deep learning optimization instructions, which are used to use convolutional neural networks to learn historical data and real-time monitoring data and automatically optimize the parameter settings of the simulation model; Cloud computing and parallel computing optimization instructions are used to perform parallel computing through the cloud computing platform, distribute simulation tasks to multiple computing nodes, and ensure the rapid processing of large-scale computing tasks.

Images