Deep learning-based intelligent identification method and system for rock and soil mechanical parameters

Abstract

The invention discloses a deep learning-based intelligent identification method and system for rock and soil mechanical parameters, and aims to accurately identify the elastic modulus, strength parameter and permeability coefficient of rock and soil and support the decision and design of engineering projects under complex geological conditions. And starting from data demand analysis, determining a parameter identification target and planning a field test layout. A basic data set is constructed by comprehensively collecting and integrating multi-source data including laboratory and field test data, remote sensing information and the like. The data is pre-processed and analyzed to remove noise and explore associations between the data. The method comprises the following core steps: constructing a rock-soil mechanical model by using a deep learning algorithm, and realizing high-precision inversion of parameters by defining and optimizing an objective function. And verifying an inversion result and carrying out uncertainty analysis. According to the method, an analysis result is fed back to engineering design adjustment and decision support, and scheme optimization is achieved. The model and strategy are dynamically adjusted through continuous monitoring and real-time data, and safety and high efficiency of engineering implementation are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of geotechnical mechanics parameter analysis, and particularly relates to an intelligent identification method and system for geotechnical mechanics parameters based on deep learning. Background Art

[0002] Geotechnical mechanics parameters such as elastic modulus, strength parameters, and permeability coefficient are affected by various factors, such as soil layer structure, particle size composition, water content, stress state, etc. There are complex non-linear relationships among these factors. Traditional empirical formulas or simple statistical methods are difficult to accurately describe this complexity. Moreover, the acquisition cost of high-quality geotechnical test data is high and the process is complex, resulting in relatively limited data volume, and there may be noise and inconsistencies in the data, affecting the accuracy and generalization ability of the model. Geotechnical engineering design and safety assessment highly rely on the accurate estimation of geotechnical mechanics parameters. Traditional methods often based on simplified assumptions are difficult to achieve high-precision prediction, especially when facing complex geological conditions. The uncertainties in geotechnical engineering come from many aspects, including the uncertainty of test data, the approximation of model assumptions, etc. Effective methods are needed to quantify and manage these uncertainties to improve the reliability of engineering decisions. This method, through the application of deep learning technology, can automatically learn complex non-linear patterns from a large amount of multi-source data, improving the accuracy and efficiency of parameter identification; at the same time, combined with optimization algorithms and uncertainty analysis, it makes engineering design and decision-making more scientific and flexible, helps to reduce engineering risks, improve construction efficiency and safety, and meet the requirements of modern geotechnical engineering projects for intelligent and refined management. Summary of the Invention

[0003] An intelligent identification method for geotechnical mechanics parameters based on deep learning includes the following steps; S1. Project planning and data requirement analysis: clarify the engineering objectives and the requirements for parameter identification: the elastic modulus, strength parameters, and permeability coefficient of the rock and soil, and plan the on-site test and monitoring arrangements, including boreholes, in-situ tests: static cone penetration test, standard penetration test, and the installation of remote monitoring sensors; S2. Data collection and integration: collect various types of on-site test data, including laboratory test data, in-situ test data, remote sensing and ground penetrating radar data, integrate historical engineering data and regional geological information, and establish a basic data set; S3. Preprocessing and data analysis: clean the collected data set, remove outliers, perform data transformation and standardization processing, and use statistical analysis to preliminarily explore the correlation between data, preparing for subsequent modeling; S4. Construct an inversion model and an objective function: use deep learning algorithms to establish a geotechnical mechanics model and integrate to obtain the formula of the geotechnical mechanics prediction model , where σrepresents stress, 𝜂 is the viscosity coefficient, which reflects the strength of the material flow or viscosity effect and is related to the flow resistance of the material, and 𝐸 is the elastic modulus, which describes the stiffness of the material within the elastic range. Calculate, where ν is Poisson's ratio, which represents the negative ratio of the transverse strain to the longitudinal strain of the material within the elastic range, reflecting the degree of transverse expansion or contraction of the material when subjected to force, G is the shear modulus, which measures the ability of the material to resist shear deformation under tangential load, 𝜀 is the strain, , Respectively represent the rate of change of stress and strain with time, that is, stress rate and strain rate, F is the yield condition, which defines the critical point where the material changes from elastic state to plastic state. The comprehensive model formula is used to predict deformation, considering the mechanical response under various working conditions, providing support for geotechnical engineering design and safety assessment, and defining the objective function according to the actual engineering situation. The objective function is minimized to reflect the error between the model prediction value and the measured value; S5. Parameter inversion and optimization: Apply the selected intelligent algorithm to perform parameter inversion, iteratively optimize the objective function to obtain the best parameter set, and perform multiple iterations and algorithm parameter adjustments to ensure convergence and parameter accuracy; S6. Result verification and uncertainty analysis: Apply the inverted parameters to the model, verify the model's prediction ability through simulation prediction and comparison with independent observation data, conduct sensitivity analysis and uncertainty assessment, understand the impact of parameter changes on model prediction, and evaluate the reliability of the results; S7. Feedback design and decision support: adjust the engineering design plan according to the inversion results, optimize the structural design and construction plan, provide risk assessment reports and suggestions, and provide scientific basis for decision makers; S8. Continuous monitoring and dynamic adjustment: During the project implementation, key parameters are continuously monitored, real-time data is used to dynamically calibrate and optimize the model, intelligent management and control are implemented, and design and construction strategies are adjusted according to the latest data feedback to ensure the safety and efficient progress of the project.

[0004] Furthermore, a deep learning-based intelligent identification method for geotechnical parameters is proposed. The specific steps of using the deep learning algorithm to establish the geotechnical mechanics model in step S4 are as follows: S41. Data preparation and preprocessing Collect data: collect historical data including stress, strain, time variables, temperature, moisture content, and experimental data. Data preprocessing: clean data, handle missing values ​​and outliers, standardize or normalize input features, and ensure data quality; S42. Feature Engineering: According to the mechanical model, construct features including strain rate, stress rate, elastic modulus, Poisson's ratio, shear modulus, and viscosity coefficient, as well as external environmental factors. Select the features crucial for model prediction through correlation analysis and principal component analysis methods; S43. Build a deep learning model: Divide the dataset into a training set, a validation set, and a test set. Use the training set to train the geomechanics model through the multi-layer perceptron algorithm, adjust the model parameters through the gradient descent optimization algorithm, and integrate to obtain the geomechanics model formula , where σ represents stress, 𝜂 is the viscosity coefficient, which reflects the strength of the material's flow or viscous effect and is related to the flow resistance of the material, 𝐸 is the elastic modulus, which describes the stiffness of the material within the elastic range, and is calculated by the formula , where ν is Poisson's ratio, which represents the negative ratio of the lateral strain to the longitudinal strain of the material within the elastic range, G is the shear modulus, 𝜀 is the strain, , respectively represent the change rates of stress and strain with time, F is the yield condition, and evaluate the model performance on the validation set. Adjust the hyperparameters such as the learning rate and batch size through methods such as cross-validation to avoid overfitting and improve the generalization ability; S44. Testing and Evaluation: Use the test set to evaluate the final performance of the model, focus on the evaluation indicators of prediction accuracy, RMSE, and R² score, analyze the sensitivity of the model to different input variables, and understand which factors have the greatest impact on the prediction results; S45. Prediction and Application Condition Simulation: Use the model formula to input parameters under different working conditions: loading conditions, changes in environmental factors, predict the mechanical responses of the rock and soil: stress, strain. According to the prediction results, evaluate the stability and safety of the geotechnical engineering under specific conditions, and continuously optimize and adjust the model based on new data in actual engineering applications to ensure the long-term effectiveness and accuracy of the model, providing a scientific basis for design optimization, risk management, and construction plans.

[0005] Furthermore, a method for intelligent identification of geomechanics parameters based on deep learning, In step S5, the PSO algorithm is used for inverse analysis of geomechanics model parameters, and the specific steps are as follows: S51. Parameter Space Definition and Constraint Condition Setting: Determine the reasonable range of each geomechanics parameter, and based on the on-site test data, set the upper and lower bounds of the parameters to ensure that the parameter combinations within the search space are physically feasible; S52. Objective Function Construction: Define an objective function to measure the degree of agreement between the model prediction results and the measured data, mean square error, mean absolute error, correlation coefficient; S53. PSO Parameters and Initial Population Setting: Considering the characteristics of geotechnical mechanics problems, finely adjust the control parameters of PSO: inertia weight and acceleration factors, to balance the global exploration and local exploitation capabilities. The positions of the initial population are randomly distributed within the parameter space, and a biased initialization is carried out based on known engineering experience and preliminary test results to accelerate convergence; S54. Model Solving and Fitness Evaluation Loop: Perform iterative calculations. For each particle, use the geotechnical mechanics model to calculate the predicted responses under its parameters: displacements and stress distributions. Compare the model prediction results with on-site monitoring and experimental data, calculate the objective function value as the fitness index, update the velocity and position according to the PSO algorithm, and at the same time consider the potential correlations between geotechnical mechanics parameters and adopt an adaptive weight adjustment strategy; S55. Result Verification and Sensitivity Analysis: Conduct a detailed analysis of the optimal solution, including sensitivity analysis, check the impact of small changes in key parameters on the model output, and use cross-validation or other statistical methods to evaluate the generalization ability of the model; S56. Application and Feedback of Parameter Inversion Results: Apply the optimized parameters back to the geotechnical mechanics model for detailed engineering analysis: stability assessment and seepage simulation. According to the model prediction results, adjust the design scheme, evaluate the construction risks, and put forward improvement suggestions.

[0006] Furthermore, an intelligent identification system for geotechnical mechanics parameters based on deep learning, the intelligent identification system for geotechnical mechanics parameters based on deep learning is used to implement any one of the intelligent identification methods for geotechnical mechanics parameters based on deep learning; the intelligent identification system for geotechnical mechanics parameters based on deep learning includes: a data requirement analysis module, a data collection and integration module, a model construction and optimization module, a result verification module, a decision support and feedback module, and a monitoring and dynamic adjustment module; Among them, the data requirement analysis module: According to the engineering objectives, clarify the geotechnical mechanics parameters to be identified (such as elastic modulus, strength parameters, permeability coefficient, etc.), and design the on-site test scheme, including borehole layout, in-situ tests (such as static cone penetration test, standard penetration test) and the layout plan of remote monitoring sensors; The data collection and integration module: Perform on-site tests and laboratory analyses, collect various types of data, integrate historical data and geological information, construct a basic database, and carry out data cleaning, outlier removal, data format unification, standardization processing, conduct preliminary statistical analysis, and explore the relationships between data; The model construction and optimization module: Construct a geotechnical mechanics model based on deep learning algorithms, define the objective function reflecting the prediction error to provide a basis for model training, apply intelligent optimization algorithms to invert the model parameters, iteratively optimize the objective function, and find the optimal parameter set; Result verification module: Use an independent dataset to verify the model's prediction ability, conduct sensitivity analysis and uncertainty assessment, and evaluate the model's reliability and parameter stability; Decision support and feedback module: Adjust the engineering design according to the model results, provide risk assessment and decision-making suggestions, and formulate an optimized construction plan; Monitoring and dynamic adjustment module: Implement continuous monitoring during the engineering process, use real-time data to dynamically correct the model, and adjust the construction strategy according to the latest situation.

[0007] Advantages of the present invention: The deep learning algorithm can learn complex data features and non-linear relationships. Compared with traditional methods, it can more accurately predict geotechnical mechanical parameters such as elastic modulus, strength, and permeability, thereby improving the accuracy of engineering design and analysis. The system can conduct sensitivity analysis and uncertainty assessment, help identify key parameters and their impacts on engineering stability, provide scientific support for risk management and decision-making, and reduce the risk of engineering failure. The real-time monitoring and model dynamic correction functions ensure the adaptability and flexibility of the project during implementation, can respond promptly to changes in geological conditions, take preventive measures, and avoid potential problems.

[0008] The automated and intelligent analysis process significantly shortens the time from data collection to decision-making, improves project management efficiency, and makes decisions more quickly and accurately. Brief description of the drawings

[0009] Figure 1 It is a flow chart of an intelligent identification method for geotechnical mechanical parameters based on deep learning; Detailed implementation manners

[0010] An intelligent identification method for geotechnical mechanical parameters based on deep learning includes the following steps; S1. Project planning and data requirement analysis: Define the engineering objectives and the requirements for parameter identification: the elastic modulus, strength parameters, and permeability coefficient of the rock and soil, and plan the on-site testing and monitoring arrangements, including boreholes, in-situ tests: static cone penetration test, standard penetration test, and installation of remote monitoring sensors; S2. Data collection and integration: Collect various on-site test data, including laboratory test data, in-situ test data, remote sensing and ground penetrating radar data, integrate historical engineering data and regional geological information, and establish a basic dataset; S3. Preprocessing and data analysis: Clean the collected dataset, remove outliers, perform data transformation and standardization processing, and use statistical analysis to preliminarily explore the correlation between data to prepare for subsequent modeling; S4. Build an inversion model and objective function: Use a deep learning algorithm to establish a geotechnical mechanics model and integrate to obtain the geotechnical mechanics prediction model formula , whereσ represents stress, 𝜂 is the viscosity coefficient, which reflects the strength of the material flow or viscosity effect and is related to the flow resistance of the material, and 𝐸 is the elastic modulus, which describes the stiffness of the material within the elastic range. Calculate, where ν is Poisson's ratio, which represents the negative ratio of the transverse strain to the longitudinal strain of the material within the elastic range, reflecting the degree of transverse expansion or contraction of the material when subjected to force, G is the shear modulus, which measures the ability of the material to resist shear deformation under tangential load, 𝜀 is the strain, , Respectively represent the rate of change of stress and strain with time, that is, stress rate and strain rate, F is the yield condition, which defines the critical point where the material changes from elastic state to plastic state. The comprehensive model formula is used to predict deformation, considering the mechanical response under various working conditions, providing support for geotechnical engineering design and safety assessment, and defining the objective function according to the actual engineering situation. The objective function is minimized to reflect the error between the model prediction value and the measured value; S5. Parameter inversion and optimization: Apply the selected intelligent algorithm to perform parameter inversion, iteratively optimize the objective function to obtain the best parameter set, and perform multiple iterations and algorithm parameter adjustments to ensure convergence and parameter accuracy; S6. Result verification and uncertainty analysis: Apply the inverted parameters to the model, verify the model's prediction ability through simulation prediction and comparison with independent observation data, conduct sensitivity analysis and uncertainty assessment, understand the impact of parameter changes on model prediction, and evaluate the reliability of the results; S7. Feedback design and decision support: adjust the engineering design plan according to the inversion results, optimize the structural design and construction plan, provide risk assessment reports and suggestions, and provide scientific basis for decision makers; S8. Continuous monitoring and dynamic adjustment: During the project implementation, key parameters are continuously monitored, real-time data is used to dynamically calibrate and optimize the model, intelligent management and control are implemented, and design and construction strategies are adjusted according to the latest data feedback to ensure the safety and efficient progress of the project.

[0011] Furthermore, a deep learning-based intelligent identification method for geotechnical parameters is proposed. The specific steps of using the deep learning algorithm to establish the geotechnical mechanics model in step S4 are as follows: S41. Data preparation and preprocessing Collect data: collect historical data including stress, strain, time variables, temperature, moisture content, and experimental data. Data preprocessing: clean data, handle missing values ​​and outliers, standardize or normalize input features, and ensure data quality; S42. Feature Engineering: According to the mechanical model, construct features including strain rate, stress rate, elastic modulus, Poisson's ratio, shear modulus, and viscosity coefficient, as well as external environmental factors. Select the features crucial for model prediction through correlation analysis and principal component analysis methods; S43. Construct a deep learning model: Divide the dataset into a training set, a validation set, and a test set. Use the training set to train the geomechanics model through the multi-layer perceptron algorithm, adjust the model parameters through the gradient descent optimization algorithm, and integrate to obtain the geomechanics model formula , where σ represents stress, 𝜂 is the viscosity coefficient, which reflects the strength of the material flow or viscous effect and is related to the flow resistance of the material. 𝐸 is the elastic modulus, which describes the stiffness of the material within the elastic range. It is calculated through the formula , where ν is Poisson's ratio, which represents the negative ratio of the transverse strain to the longitudinal strain of the material within the elastic range. G is the shear modulus, 𝜀 is the strain, , respectively represent the change rates of stress and strain with time. F is the yield condition, and evaluate the model performance on the validation set. Adjust the hyperparameters such as the learning rate and batch size through methods such as cross-validation to avoid overfitting and improve the generalization ability; S44. Testing and Evaluation: Use the test set to evaluate the final performance of the model. Pay attention to the evaluation indicators such as prediction accuracy, RMSE, and R² score, analyze the sensitivity of the model to different input variables, and understand which factors have the greatest impact on the prediction results; S45. Prediction and Application Condition Simulation: Use the model formula to input parameters under different working conditions: loading conditions, changes in environmental factors, and predict the mechanical responses of the rock and soil: stress and strain. According to the prediction results, evaluate the stability and safety of geotechnical engineering under specific conditions, and continuously optimize and adjust the model based on new data in actual engineering applications to ensure the long-term effectiveness and accuracy of the model, providing a scientific basis for design optimization, risk management, and construction plans.

[0012] Furthermore, an intelligent identification method for geomechanics parameters based on deep learning, In step S5, the PSO algorithm is used for the inversion of geomechanics model parameters. The specific steps are as follows: S51. Parameter Space Definition and Constraint Condition Setting: Determine the reasonable range of each geomechanics parameter. Based on the on-site test data, set the upper and lower bounds of the parameters to ensure that the parameter combinations within the search space are physically feasible; S52. Objective Function Construction: Define an objective function to measure the degree of agreement between the model prediction results and the measured data, including mean square error, mean absolute error, and correlation coefficient; S53. PSO Parameters and Initial Population Setting: Considering the characteristics of geotechnical mechanics problems, finely adjust the control parameters of PSO: inertia weight and acceleration factors, to balance the global exploration and local exploitation capabilities. The positions of the initial population are randomly distributed within the parameter space, and are initialized with bias based on known engineering experience and preliminary test results to accelerate convergence; S54. Model Solving and Fitness Evaluation Loop: Perform iterative calculations. For each particle, use the geotechnical mechanics model to calculate the predicted responses under its parameters: displacement and stress distribution. Compare the model prediction results with on-site monitoring and experimental data, calculate the objective function value as the fitness index, update the velocity and position according to the PSO algorithm, and adopt an adaptive weight adjustment strategy while considering the potential correlations among geotechnical mechanics parameters; S55. Result Verification and Sensitivity Analysis: Conduct a detailed analysis of the optimal solution, including sensitivity analysis, to examine the impact of small changes in key parameters on the model output, and use cross-validation or other statistical methods to evaluate the generalization ability of the model; S56. Application and Feedback of Parameter Inversion Results: Apply the optimized parameters back to the geotechnical mechanics model for detailed engineering analysis: stability assessment and seepage simulation. According to the model prediction results, adjust the design plan, evaluate the construction risks, and propose improvement suggestions.

[0013] An intelligent identification system for geotechnical mechanics parameters based on deep learning, where the intelligent identification system for geotechnical mechanics parameters based on deep learning is used to implement any one of the intelligent identification methods for geotechnical mechanics parameters based on deep learning; the intelligent identification system for geotechnical mechanics parameters based on deep learning includes: a data requirement analysis module, a data collection and integration module, a model construction and optimization module, a result verification module, a decision support and feedback module, and a monitoring and dynamic adjustment module; Among them, the data requirement analysis module: According to the engineering objectives, clarify the geotechnical mechanics parameters to be identified (such as elastic modulus, strength parameters, permeability coefficient, etc.), and design the on-site test plan, including borehole layout, in-situ tests (such as static cone penetration test, standard penetration test) and the layout plan of remote monitoring sensors; The data collection and integration module: Execute on-site tests and laboratory analyses, collect various types of data, integrate historical data and geological information, construct a basic database, and perform data cleaning, outlier removal, data format unification, standardization processing, conduct preliminary statistical analyses, and explore the relationships among the data; The model construction and optimization module: Construct a geotechnical mechanics model based on deep learning algorithms, define the objective function reflecting the prediction error to provide a basis for model training, apply intelligent optimization algorithms to invert the model parameters, iteratively optimize the objective function, and find the optimal parameter set; Result verification module: Use an independent dataset to verify the model's prediction ability, conduct sensitivity analysis and uncertainty assessment, and evaluate the model's reliability and parameter stability; Decision support and feedback module: Adjust the engineering design according to the model results, provide risk assessment and decision-making suggestions, and formulate an optimized construction plan; Monitoring and dynamic adjustment module: Implement continuous monitoring during the engineering process, dynamically correct the model using real-time data, and adjust the construction strategy according to the latest situation.

Claims

1. An intelligent identification method for geotechnical mechanics parameters based on deep learning, characterized in that, Including the following steps; S1. Project planning and data requirement analysis: Define the engineering objectives and the requirements for parameter identification: the elastic modulus, strength parameters, and permeability coefficient of the rock and soil, and plan the on-site testing and monitoring arrangements, including boreholes, in-situ tests: static cone penetration test, standard penetration test, and installation of remote monitoring sensors; S2. Data collection and integration: Collect various on-site test data, including laboratory test data, in-situ test data, remote sensing and ground penetrating radar data, integrate historical engineering data and regional geological information, and establish a basic dataset; S3. Preprocessing and data analysis: Clean the collected dataset, remove outliers, perform data transformation and standardization, and use statistical analysis to preliminarily explore the correlations between data to prepare for subsequent modeling; S4. Construct the inversion model and objective function: Use deep learning algorithms to establish a geomechanics model and integrate it to obtain the geomechanics model formula , where σ represents stress, 𝜂 is the viscosity coefficient, which reflects the strength of material flow and viscous effects and is related to the flow resistance of the material. 𝐸 is the elastic modulus, which describes the stiffness of the material within the elastic range. It is calculated by the formula , where ν is the Poisson's ratio, which represents the negative ratio of the transverse strain to the longitudinal strain of the material within the elastic range, G is the shear modulus, 𝜀 is the strain, , respectively represent the rates of change of stress and strain with time. F is the yield condition. Use the comprehensive model formula for deformation prediction, consider the mechanical responses under various working conditions, provide support for geotechnical engineering design and safety assessment, and define the objective function according to the actual engineering situation. Minimizing the objective function reflects the error between the model prediction value and the measured value; S5. Parameter inversion and optimization: Apply the PSO algorithm for parameter inversion of the geomechanical model, iteratively optimize the objective function to obtain the optimal parameter set, with multiple iterations and algorithm parameter adjustments; S6. Result verification and uncertainty analysis: Apply the inverted parameters to the model, verify the prediction ability of the model by comparing the simulation predictions with independent observation data, conduct sensitivity analysis and uncertainty assessment, understand the impact of parameter changes on the model predictions, and evaluate the reliability of the results; S7. Feedback design and decision support: Adjust the engineering design scheme according to the inversion results, optimize the structural design and construction plan, provide a risk assessment report and recommendations, and provide a scientific basis for decision-makers; S8. Continuous monitoring and dynamic adjustment: Continuously monitor key parameters during the engineering implementation process, use real-time data for dynamic calibration and optimization of the model, implement intelligent control, and adjust the design and construction strategies according to the latest data feedback to ensure the safe and efficient progress of the project.

2. A method for intelligent identification of geomechanical parameters based on deep learning as claimed in claim 1, wherein The specific steps of establishing a geomechanical model using a deep learning algorithm in step S4 are as follows: S41. Data preparation and preprocessing - data collection: Collect historical data, experimentally obtained data containing stress, strain, time variables, temperature, and moisture content, data preprocessing: Clean the data, handle missing values and outliers, standardize or normalize the input features to ensure data quality; S42. Feature engineering: According to the mechanical model, construct features including strain rate, stress rate, elastic modulus, Poisson's ratio, shear modulus, and viscosity coefficient, as well as external environmental factors, and screen out the features crucial for model prediction through correlation analysis and principal component analysis methods; S43. Construct a deep learning model: Divide the dataset into a training set, a validation set, and a test set. Use the training set to train the geomechanics model through the multi-layer perceptron algorithm, adjust the model parameters through the gradient descent optimization algorithm, and integrate to obtain the geomechanics model formula , where σ represents stress, 𝜂 is the viscosity coefficient, which reflects the strength of material flow and viscous effects and is related to the flow resistance of the material. 𝐸 is the elastic modulus, which describes the stiffness of the material within the elastic range and is calculated by the formula , where ν is the Poisson's ratio, which represents the negative ratio of the transverse strain to the longitudinal strain of the material within the elastic range, G is the shear modulus, 𝜀 is the strain , respectively represent the rates of change of stress and strain with time, F is the yield condition, and evaluate the model performance on the validation set. Adjust the hyperparameters, such as the learning rate and batch size, through methods such as cross-validation to avoid overfitting and improve the generalization ability; S44. Testing and evaluation: Use the test set to evaluate the final performance of the model, focus on evaluation metrics such as prediction accuracy, RMSE, and R² score, analyze the sensitivity of the model to different input variables, and understand which factors have the greatest impact on the prediction results; S45. Prediction and application condition simulation: Input parameters under different working conditions into the model formula: loading conditions and changes in environmental factors, predict the mechanical responses of rock and soil: stress and strain. According to the prediction results, evaluate the stability and safety of geotechnical engineering under specific conditions, and continuously optimize and adjust the model based on new data in actual engineering applications to ensure the long-term effectiveness and accuracy of the model, providing a scientific basis for design optimization, risk management, and construction plans.

3. A method for intelligent identification of geotechnical mechanical parameters based on deep learning as claimed in claim 1, characterized in that In step S5, the PSO algorithm is used for inverse analysis of geotechnical mechanical model parameters, and the specific steps are as follows: S51. Parameter space definition and constraint condition setting: Determine the reasonable range of each geotechnical mechanical parameter, and based on on-site test data, set the upper and lower bounds of the parameters to ensure that the parameter combinations within the search space are physically feasible; S52. Objective function construction: Define an objective function to measure the degree of agreement between the model prediction results and the measured data, including mean square error, mean absolute error, and correlation coefficient; S53. PSO parameter and initial population setting: Considering the characteristics of geotechnical mechanical problems, finely adjust the control parameters of PSO: inertia weight and acceleration factor, to balance the global exploration and local development capabilities. The positions of the initial population are randomly distributed within the parameter space, and biased initialization is performed based on known engineering experience and preliminary test results to accelerate convergence; S54. Model solution and fitness evaluation loop: Iteratively calculate. For each particle, use the geotechnical mechanical model to calculate the predicted responses under its parameters: displacement and stress distribution. Compare the model prediction results with on-site monitoring and experimental data, calculate the objective function value as the fitness index, update the velocity and position according to the PSO algorithm, and at the same time consider the potential correlation between geotechnical mechanical parameters and adopt an adaptive weight adjustment strategy; S55. Result verification and sensitivity analysis: Conduct a detailed analysis of the optimal solution, including sensitivity analysis, check the impact of small changes in key parameters on the model output, and use cross-validation or other statistical methods to evaluate the generalization ability of the model; S56. Application and feedback of the inverse analysis results of parameters: Apply the optimized parameters back to the geotechnical mechanical model for detailed engineering analysis: stability evaluation and seepage simulation. According to the model prediction results, adjust the design plan, evaluate the construction risk, and put forward improvement suggestions.

4. An intelligent identification system for geotechnical mechanics parameters based on deep learning, characterized in that, The method for intelligent identification of geotechnical mechanical parameters based on deep learning is used to implement the method for intelligent identification of geotechnical mechanical parameters based on deep learning as described in any one of claims 1-3; The system for intelligent identification of geotechnical mechanical parameters based on deep learning includes: a data requirement analysis module, a data collection and integration module, a model construction and optimization module, a result verification module, a decision support and feedback module, and a monitoring and dynamic adjustment module; Among them, the data requirement analysis module: According to the engineering objectives, clarify the geotechnical mechanical parameters to be identified, and design an on-site test plan, including borehole layout, in-situ test, and layout plan of remote monitoring sensors; Data Acquisition and Integration Module: Conduct on-site tests and laboratory analyses, collect various types of data, integrate historical data and geological information, construct a basic database, and perform data cleaning, outlier removal, data format unification, standardization processing, conduct preliminary statistical analyses, and explore the relationships between data; Model Construction and Optimization Module: Construct a geomechanical model based on deep learning algorithms, define an objective function reflecting the prediction error to provide a basis for model training, apply intelligent optimization algorithms to invert the model parameters, iteratively optimize the objective function, and find the optimal parameter set; Result Verification Module: Use an independent dataset to verify the model's prediction ability, conduct sensitivity analyses and uncertainty assessments, and evaluate the model's reliability and parameter stability; Decision Support and Feedback Module: Adjust the engineering design according to the model results, provide risk assessments and decision-making suggestions, and formulate optimized construction plans; Monitoring and Dynamic Adjustment Module: Implement continuous monitoring during the engineering process, use real-time data to dynamically correct the model, and adjust the construction strategy according to the latest situation.

Images