Rock-soil body parameter automatic correction method and system, electronic equipment and storage medium

Abstract

The invention relates to the technical field of geological engineering and automatic control, and discloses a rock-soil body parameter automatic correction method and system, electronic equipment and a storage medium, and the method comprises the steps: carrying out the deployment and data collection of a sensor network; carrying out data preprocessing and abnormal value detection on the collected data; applying a mechanical behavior model and correcting parameters of a rock-soil body; applying an optimization algorithm and optimizing key mechanical parameters; the feedback calibration and correction parameter adjustment system comprises a sensor module, a data processing module, an optimization module and a calibration module, and the electronic equipment comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor. According to the method, the stress and strain data of the rock-soil body are collected in real time, key mechanical parameters are corrected in combination with an optimization algorithm, high consistency of the parameters of the rock-soil body and the actual working condition is kept through real-time monitoring and dynamic correction, and the correction precision and the engineering safety are improved.

Description

Technical Field

[0001] The present invention relates to the field of geological engineering and automated control technology, and in particular to a method, system, electronic equipment and storage medium for automatic correction of rock and soil parameters. Background Art

[0002] Geotechnical engineering plays a vital role in modern construction, infrastructure development, and geological disaster prevention. In these projects, the accuracy of geotechnical parameters is fundamental to ensuring construction safety and structural stability. Geotechnical parameters such as elastic modulus, Poisson's ratio, and shear modulus not only describe the mechanical behavior of geotechnical structures but also directly influence structural design, construction plan selection, and subsequent maintenance decisions.

[0003] Existing methods for obtaining and correcting rock and soil parameters mainly rely on two means: one is traditional laboratory testing, which obtains mechanical parameters through sampling, compression testing, triaxial testing, etc.; the other is collecting rock and soil stress, strain, displacement and other data through field monitoring networks. These data are corrected through numerical models to help determine the actual bearing capacity and deformation characteristics of the rock and soil.

[0004] However, existing geotechnical parameter correction methods, which are usually based on manual data collection and analysis methods, are inefficient, have limited accuracy, and are easily interfered with by human factors. Most existing parameter correction schemes are based on static preset models. However, geotechnical bodies are affected by multiple factors during actual construction, such as changes in groundwater levels and fluctuations in construction loads. As a result, the correction results cannot timely reflect the true state of the geotechnical body, affecting the safety and progress of construction. In addition, due to the lack of a real-time feedback mechanism, the correction process cannot adjust the geotechnical parameters according to real-time data, resulting in a large lag. Therefore, the present invention provides a method, system, electronic device, and storage medium for automatic correction of geotechnical parameters to address the shortcomings of the existing technology. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a method, system, electronic device and storage medium for automatic correction of rock and soil parameters, which solves the problem that the existing rock and soil parameter correction methods are usually based on manual data collection and analysis methods with low efficiency, limited accuracy, and easily interfered by human factors. Most parameter correction schemes are based on static preset models, and the rock and soil are affected by multiple factors during the actual construction process, resulting in the correction results being unable to timely reflect the true state of the rock and soil.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for automatically correcting rock and soil parameters, comprising the following steps: Numerical simulation is used to determine the sensor installation location, and the stress and strain parameters of the rock and soil are collected in real time through the sensor network, and the collected data are transmitted to the data processing center; Preprocess the collected data, use outlier detection algorithms to detect and remove abnormal data, and use interpolation methods to repair missing data; Based on the pre-processed data, the mechanical behavior model of the rock and soil is applied to modify the key mechanical parameters of the rock and soil, including the elastic modulus and Poisson's ratio; Use optimization algorithms to optimize the key mechanical parameters of rock and soil, combine Lagrange multiplier method and variational method to minimize the correction error and meet the constraints to ensure the accuracy of the correction results; Feedback calibration is performed based on the modified key mechanical parameters of the rock and soil mass. By comparing with subsequent real-time monitoring data, the error is calculated and the modified key mechanical parameters are adjusted based on the feedback.

[0007] Preferably, the sensor installation position is determined by numerical simulation, and the numerical simulation step includes: analyzing the stress and strain distribution of the rock and soil under different working conditions by numerical simulation to identify key areas; Design sensor layout based on simulation results to ensure monitoring of key areas; The sensor installation position is optimized based on the rock and soil characteristics, deformation laws and construction conditions to ensure maximum data collection accuracy; Optimize sensor layout density to ensure comprehensive monitoring and improve data quality; Combine simulation results with actual construction to adjust sensor positions to ensure accurate monitoring.

[0008] Preferably, the data preprocessing step includes: Obtain the original monitoring data transmitted by the sensor, perform preliminary cleaning, and remove noise and irrelevant data; Use standard deviation methods and machine learning algorithms to identify and remove outliers, and perform interpolation repairs on missing or abnormal data to ensure data integrity and accuracy; Evaluate data quality to ensure that the pre-processed data meets the requirements for subsequent corrections.

[0009] Preferably, the step of performing dimensionality reduction processing on the collected data includes: Convert the collected high-dimensional data into tensor form and use tensor decomposition technology to extract the main features of the data; The principal component analysis method is used to reduce the dimensionality of the data, retaining the component with the largest variance in the data, thereby reducing redundant information; the feature selection method is used to further screen the features that are most relevant to security threat analysis.

[0010] Preferably, the optimization algorithm includes: Use Lagrange multiplier method and variational method to optimize key mechanical parameters of geotechnical mass, minimize correction error and meet engineering constraints; The correction results are further optimized through polynomial regression and least squares method to ensure the accuracy of the correction parameters and provide preliminary correction parameters for subsequent feedback calibration.

[0011] Preferably, the feedback calibration step includes: By comparing with subsequent real-time monitoring data, the correction error is calculated and the correction coefficient is adjusted. The real-time monitoring data includes displacement data, environmental data, construction progress and load changes; The correction error is evaluated using the mean square error, and the correction coefficient is dynamically adjusted through an adaptive feedback mechanism. The formula is: ΔP adjusted =ΔP initial +γ×ΔP error ; Where ΔP adjusted is the adjusted rock and soil parameter, ΔP initial is the preliminary correction value output by the optimization algorithm, and γ is the dynamic inverse constant ΔP error is the error between the actual measurement value and the optimized result.

[0012] The correction parameters are updated according to the feedback results, and multiple iterations are performed to ensure the accuracy of the final correction results.

[0013] Preferably, the modified key mechanical parameters are dynamically adjusted by feedback gain and error threshold, and the formula is: E adjusted =E new +K·(RMSE-threshold); Among them, E adjusted is the adjusted elastic modulus of rock and soil, E new is the newly corrected elastic modulus of rock and soil, K is the feedback gain, threshold is the preset error threshold, and RMSE is the root mean square error.

[0014] It also provides an electronic equipment system for automatic correction of geotechnical parameters, including: Sensor networks are used to collect stress and strain data of rock and soil in real time; The data processing unit is used to pre-process the collected data and correct the mechanical parameters of the rock and soil based on the pre-processed data; the optimization module is used to execute the optimization algorithm to optimize the mechanical parameters of the rock and soil to ensure the accuracy of the correction results; The calibration module is used to adjust the correction results according to real-time feedback to ensure the accuracy of the final correction parameters.

[0015] Also provided is an electronic device for automatic correction of rock and soil parameters, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a method for automatic correction of rock and soil parameters can be implemented.

[0016] A rock and soil mass parameter automatic correction storage medium is also provided, which stores a computer program. When the computer program is executed by a processor, the rock and soil mass parameter automatic correction method can be implemented.

[0017] The present invention provides a method, system, electronic device, and storage medium for automatically correcting rock and soil parameters. The method has the following beneficial effects: 1. The present invention adopts a technical solution combining numerical simulation with sensor networks. By collecting stress and strain data of rock and soil in real time and combining it with optimization algorithms, key mechanical parameters are corrected. Through real-time monitoring and dynamic correction, rock and soil parameters are kept highly consistent with actual working conditions. Compared with the manual collection and manual correction solutions in the existing technology, this solves the problems of data deviation and correction lag caused by human factors, and improves correction accuracy and engineering safety.

[0018] 2. The present invention ensures the accuracy and integrity of data during transmission and processing by combining multiple outlier detection algorithms with interpolation methods. It achieves the effect of effectively filtering out abnormal data and repairing missing data, ensuring that subsequent model analysis is based on reliable data. Compared with the single outlier processing method in traditional technology, it solves the problem of missing or misjudging abnormal data during data cleaning, and improves data quality and the reliability of analysis results.

[0019] 3. The present invention combines the mechanical behavior model with the optimization algorithm, and can dynamically adjust the key mechanical parameters of the rock and soil mass based on real-time monitoring data; it achieves the effect of making the corrected parameters closer to the actual working conditions through optimization and feedback calibration; compared with the fixed model and preset parameters in the prior art, the present invention solves the problem of being unable to correct the model parameters in a timely manner under complex working conditions, and greatly improves adaptability and accuracy.

[0020] 4. This invention uses a dynamic correction and multiple iterative optimization mechanism to adjust parameters after each feedback calibration. This achieves real-time optimization and correction results and can cope with parameter changes in complex construction environments. Compared with the static correction schemes in the existing technology, this solves the problem of parameter correction lag caused by environmental changes and asynchronous construction progress, significantly improving correction accuracy and reliability during the construction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flow chart of the method steps of the present invention; Figure 2 This is a system architecture diagram of the present invention; Figure 3 Schematic diagram of the computer device structure of the present invention.

[0022] Among them, 40, computer equipment; 41, processor; 42, memory; 43, storage medium. DETAILED DESCRIPTION

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0024] Please see the attached Figure 1 , an embodiment of the present invention provides a method for automatically correcting rock and soil parameters, comprising the following steps: S1. Use numerical simulation to determine the sensor installation location, collect stress and strain parameters of the rock and soil in real time through the sensor network, and transmit the collected data to the data processing center; S2. Preprocess the collected data, use an outlier detection algorithm to detect and remove abnormal data, and use interpolation to repair missing data; S3. Based on the preprocessed data, apply the mechanical behavior model of the rock and soil to modify the key mechanical parameters of the rock and soil, including the elastic modulus and Poisson's ratio; S4. Use optimization algorithms to optimize the key mechanical parameters of the rock and soil mass, combining the Lagrange multiplier method and the variational method to minimize the correction error and meet the constraints to ensure the accuracy of the correction results; S5. Feedback calibration is performed based on the modified key mechanical parameters of the rock and soil mass. By comparing with subsequent real-time monitoring data, the error is calculated and the modified key mechanical parameters are adjusted based on the feedback.

[0025] In step S1, in this embodiment, numerical simulation is used to analyze the distribution characteristics of the rock mass, such as stress and strain, and the sensor installation locations are selected based on the physical properties of the target rock mass, construction conditions, and engineering requirements. Numerical simulation can help identify potential risk areas and critical areas within the rock mass, ensuring that the sensor deployment covers all critical locations. Through simulation calculations, information such as the stress distribution, deformation trends, and potential slip surfaces of the rock mass under different working conditions can be obtained, providing theoretical support for the optimal sensor placement.

[0026] Specifically, numerical simulation of geotechnical materials uses mechanical models to comprehensively analyze the geotechnical material. Numerical simulation techniques such as the finite element method (FEM) or discrete element method (DEM) are used to simulate the mechanical behavior of the geotechnical material under different loading conditions. By inputting the geotechnical material's physical and mechanical parameters (such as elastic modulus and Poisson's ratio) and boundary conditions, the stress and strain states in different regions are calculated.

[0027] Alternatively, numerical simulations can be used to further optimize sensor placement by performing a detailed zoning analysis of the stress field in the rock and soil mass, incorporating the mechanical parameters of different soil layers. This strategy allows for a denser sensor deployment in key areas (such as the perimeter of the foundation pit, tunnel surrounding rock, and slopes, where deformation is likely) to ensure more accurate monitoring of deformation in these areas.

[0028] In one possible implementation, the system automatically generates an optimal sensor layout plan through an algorithm, taking into account various construction scenarios, natural disasters, and specific project requirements. This plan not only considers the distribution of stress and strain, but also practical factors such as communication distance between sensors, energy consumption, and installation ease.

[0029] Furthermore, during implementation, the sensor locations determined by the system through numerical simulations will be applied to the actual construction site. Typically, sensor locations are dynamically adjusted based on changes during the construction process to respond to new monitoring needs or abnormal situations. For example, during foundation pit construction, areas of concentrated stress may appear. In these cases, the system can use numerical simulations to reposition and increase the density of sensors, ensuring accurate monitoring of these critical areas.

[0030] Furthermore, sensor networks must also possess excellent real-time performance and efficient transmission capabilities when collecting data. Wireless ad hoc networking technology can be used to establish stable communication links through collaboration between sensor nodes without external network support. In some embodiments, low-power, high-reliability wireless communication protocols such as LoRa and ZigBee can be selected. These protocols can ensure stable data transmission while reducing energy consumption.

[0031] Typically, sensor networks collect data in real time, transmitting raw data to a data processing center with millisecond latency. This undisturbed data collection process ensures all critical information is fed back in real time, providing raw data for subsequent preprocessing and correction.

[0032] Specifically, sensors continuously monitor stress and strain data within the geotechnical mass and transmit this data to a data receiving center at a set sampling frequency. There, the data is collated, stored, and synchronized based on the transmitted timestamp. This data includes, but is not limited to, geotechnical parameters such as stress, strain, displacement, and temperature, and supports subsequent correction calculations.

[0033] In this embodiment, in some specific application scenarios, the installation of sensors is not limited to traditional soil. Special areas such as rock slopes and underground structures (such as tunnels and foundation pits) can also be considered. For these areas, the layout of sensors will be more complex, usually taking into account the monitoring needs of different depths and layers.

[0034] Regarding step S2, in this embodiment, after data collection is complete, the raw data is first preliminarily cleaned to remove noise and irrelevant data to ensure data accuracy. The primary goal of data cleaning is to remove outliers from the monitoring data to ensure that subsequent processing is not disrupted. To achieve this goal, the present invention employs multiple outlier detection methods to ensure that all types of abnormal data can be effectively identified and processed.

[0035] In general, the data preprocessing step first uses the standard deviation method to detect outliers in the data. Specifically, during the monitoring process, for each parameter collected by the sensor (such as stress, strain, displacement, etc.), according to statistical principles, if a data point exceeds the range of three times the standard deviation of the mean, the data point is considered an outlier. Through this method, data points that are significantly deviated from the normal range can be identified relatively quickly and effectively. The specific formula is as follows: |x i -μ|>3σ; Among them, x i is the i-th data point, μ is the mean of the data, and σ is the standard deviation of the data. If the deviation of a data point exceeds three times the standard deviation, the point is marked as an outlier.

[0036] Alternatively, the standard deviation method can be combined with the IsolationForest algorithm for further outlier detection. The IsolationForest algorithm is a tree-based algorithm that constructs multiple decision trees to determine data anomalies. This method can more effectively identify outliers in high-dimensional data, especially for rare outliers in a dataset. The algorithm determines the degree of anomaly by calculating the degree of isolation of a sample, making it easier to distinguish outliers that differ significantly from other data points.

[0037] Specifically, when performing outlier detection, IsolationForest performs multiple random partitions on the dataset and calculates the degree of isolation of each data point. The degree of isolation determines whether a data point is an outlier. If a data point exhibits a high degree of isolation in most trees, it is considered an outlier. This method is particularly suitable for data with multiple dimensions and complex distributions.

[0038] In one possible implementation, linear interpolation can be used to repair data points that are determined to be abnormal. Linear interpolation is a common data interpolation method that estimates and fills in missing or abnormal data based on the trends of adjacent data points. Specifically, if the data value of a sensor is determined to be abnormal, the interpolation formula can be used based on the data values of the sensor at adjacent time points: Among them, x1 and x2 are the horizontal coordinates of the known data points, y1 and y2 are the vertical coordinates of the known data points, x is the horizontal coordinate of the interpolation point, and y is the interpolation result. Through this method, the gaps of abnormal data can be accurately filled to ensure the continuity of the data.

[0039] Additionally, you can choose to directly remove detected outliers. After removing the outliers, the system recalculates the mean and standard deviation of the remaining data to ensure the cleanliness and validity of the dataset. This approach is suitable for outlier data that cannot be restored through interpolation, preventing the impact of erroneous data on subsequent calculations and model revisions.

[0040] In this embodiment, in order to further improve the data quality, a data quality assessment and feedback mechanism is adopted. After completing the outlier cleaning, the system will perform a quality assessment on the pre-processed data. The evaluation criteria mainly include aspects such as data accuracy, completeness and consistency. Specifically, data accuracy refers to the degree of deviation between the data and the actual monitoring value; data integrity refers to the degree of data loss or damage; data consistency refers to the consistency between data from different sensors. By establishing an evaluation indicator system, the system can regularly generate data quality reports and provide timely feedback to monitoring personnel and data analysis teams. If the evaluation results show that the data quality is poor, the system will trace the data collection and transmission process to find and fix possible problems.

[0041] Specifically, during the data quality assessment process, the system evaluates the accuracy of the data based on the following formula: Among them, Accuracy represents the average error between the predicted result and the actual value. is the predicted rock and soil parameter, yi is the actual measured geotechnical parameter, and n is the number of data points. This formula can be used to quantify the error between the preprocessed data and the actual monitoring data, and further evaluate the data quality.

[0042] In step S3, in this embodiment, after data preprocessing and outlier detection, the geotechnical mechanical behavior model is applied to correct key mechanical parameters of the geotechnical material. Based on the cleaned and interpolated monitoring data, the mechanical behavior model is used to calculate key mechanical parameters of the geotechnical material, including but not limited to elastic modulus and Poisson's ratio. By correcting these mechanical parameters, the geotechnical model can more accurately reflect the behavior of the geotechnical material under actual working conditions, thereby providing precise input for subsequent optimization algorithms and feedback calibration.

[0043] In this example, once the data preprocessing steps are complete, the resulting clean data is used as input to a mechanical behavior model for the geotechnical mass. This model describes the stress and strain response of the geotechnical mass under external loads. Typically, the behavior of geotechnical mass can be represented by classical constitutive models, such as elastic and plastic models, as well as more complex nonlinear constitutive models.

[0044] Specifically, the present invention uses nonlinear constitutive models suitable for geotechnical materials, such as the Modified Cam-Clay model or the Bouc-Wen model, to better simulate the deformation and rheological properties of geotechnical materials under different loads. These models connect the relationship between stress and strain through mechanical equations and can accurately describe the physical behavior of geotechnical materials under complex working conditions. The specific model can be expressed as: σ=f(ε,κ,p); Here, σ is the stress in the geotechnical mass, ε is the strain, κ is the plastic strain, and p is the pore pressure. The function f can represent different constitutive relations. By adjusting these parameters, the changing behavior of the geotechnical mass under different working conditions can be reflected. The accuracy of the mechanical model is crucial, especially when considering the complexities of the engineering environment (such as groundwater infiltration and construction disturbances).

[0045] Alternatively, in certain embodiments, the mechanical behavior model used can be further refined into a multi-layered geological structure model to simulate the mechanical behavior of different soil layers. Depending on the different layers of the rock mass (e.g., sand, clay, rock, etc.), the constitutive model selected will vary. For clay layers, a model based on the Mohr-Coulomb criterion can be selected to describe its plastic behavior, while for sand layers, a more simplified elastic constitutive model can be used.

[0046] Generally, during the parameter correction process, optimization algorithms are used to further ensure the accuracy and rationality of the correction results. Optimization algorithms can minimize correction errors and ensure that the corrected geotechnical parameters meet various engineering constraints. In this embodiment, the optimization steps in the correction process utilize the Lagrange multiplier method and the variational method. Combining these algorithms effectively addresses constraints and optimizes the objective function, ensuring that the correction results both conform to physical laws and meet the needs of engineering practice.

[0047] For example, when applying the Lagrange multiplier method, the optimization objective can be set to minimize the correction error, as follows: in, is the optimization objective function, which represents the target of the correction parameter, E is the elastic modulus of the rock and soil, ν is the Poisson's ratio of the rock and soil, σ measured is the measured stress value, σ predicted is the stress value calculated based on the model, λ is the Lagrange multiplier, and constraint conditions represent the physical limitations or engineering requirements in the optimization problem. The constraint conditions include the actual deformation capacity of the rock and soil mass and the experimental test conditions.

[0048] Alternatively, in some embodiments, if the correction results for certain parameters exceed the set range, the system automatically adjusts the correction algorithm and performs multiple iterations until the correction results meet the required accuracy. This iterative optimization process can effectively address parameter inaccuracies caused by external interference or changes in the construction environment, ensuring the accuracy of the geotechnical parameter correction results.

[0049] In step S4, in this embodiment, key parameters of the geotechnical mass (such as elastic modulus and Poisson's ratio) are precisely adjusted to ensure the accuracy of the correction results and the feasibility of the project. The optimization algorithm further improves the accuracy of the model through repeated adjustments during the correction process.

[0050] In this embodiment, the optimization algorithm primarily uses a combination of the Lagrange multiplier method and the calculus of variations to optimize the key mechanical parameters of the geotechnical mass. After applying the geotechnical mechanical behavior model, the resulting parameter corrections may still contain errors, necessitating further corrections through the optimization algorithm. Specifically, the optimization algorithm adjusts itself based on the objective function to minimize the correction errors and ensure that the corrections meet the engineering constraints.

[0051] In general, the Lagrange multiplier method is used to deal with constrained optimization problems. The Lagrange multiplier method is a common method for dealing with constrained optimization problems. Its basic idea is to transform the constraints into part of the objective function, construct the Lagrange function, and find the optimal solution by taking the derivative of the Lagrange function. For the optimization of geotechnical parameters, the objective function can be expressed as the sum of squares of the correction errors, that is, minimizing the correction errors so that the corrected geotechnical parameters are closer to the true values. The optimization objective function can be expressed as: in, is the optimization objective function, which represents the target of the correction parameter, E is the elastic modulus of the rock and soil, ν is the Poisson's ratio of the rock and soil, σ measured is the measured stress value, σ predicted is the stress value calculated based on the model, λ is the Lagrangian multiplier, and constraint conditions represent the physical limitations or engineering requirements in the optimization problem. The constraint conditions include the actual deformation capacity of the rock and soil mass and the experimental test conditions. By optimizing the Lagrangian function, the optimal correction value of the rock and soil mass parameters can be obtained, thereby ensuring the accuracy of the model.

[0052] Specifically, during the optimization process, the Lagrange multiplier method minimizes the correction error of the mechanical parameters of the geotechnical mass, while taking into account the actual constraints in the working conditions. During the correction process of the elastic modulus and Poisson's ratio of the geotechnical mass, the optimization algorithm adjusts the values of these two parameters based on the stress and strain data in the actual working conditions to meet the engineering design requirements and actual physical limitations.

[0053] Alternatively, to enhance the stability and accuracy of the optimization algorithm, the calculus of variations can be incorporated for correction. The calculus of variations is a mathematical method used to solve optimization problems, seeking the optimal solution to the objective function by minimizing variations. In this invention, the calculus of variations is used to further optimize the correction errors, ensuring that the corrected geotechnical parameters meet practical requirements in engineering applications.

[0054] Specifically, the calculus of variations optimizes the objective function using the variational principle, effectively handling optimization problems with multiple constraints. The goal of the optimization process is to minimize the correction error while satisfying a series of physical and engineering constraints. The application of the calculus of variations enables the optimization process to more accurately reflect the actual state of the geotechnical mass and provide real-time parameter corrections in a constantly changing construction environment.

[0055] In one possible implementation, the iterative process of the optimization algorithm is adjusted by the following formula: E adjusted =E model +ΔE; ν adjusted =νmodel +Δν; Among them, E adjusted is the adjusted elastic modulus of rock and soil, ν adjusted is the adjusted Poisson's ratio, E model and ν model They represent the initial elastic modulus and Poisson's ratio based on the model, respectively. ΔE and Δν are correction values obtained through the optimization algorithm. The optimization algorithm minimizes the prediction error by repeatedly calculating and adjusting the correction values, making the corrected geotechnical parameters closer to the actual values.

[0056] In some embodiments, to further improve the efficiency of the optimization algorithm, heuristic algorithms (such as genetic algorithms and simulated annealing algorithms) can be used to assist the optimization process. These algorithms can more quickly search for the global optimal solution, avoid being trapped in local optimal solutions, and thus accelerate the entire parameter optimization process. During the construction process, the constant changes in real-time data may require rapid adjustment of geotechnical parameters. Therefore, the use of heuristic optimization algorithms can significantly improve the system's response speed.

[0057] In step S5, in this embodiment, the correction error is calculated by comparing the real-time monitoring data with the optimization results, and the correction parameters are dynamically adjusted based on the error feedback. The core goal of this step is to ensure that the corrected geotechnical parameters match the actual working conditions, thereby improving the accuracy of the correction results and ensuring that the geotechnical parameters are always highly consistent with the actual project requirements.

[0058] Typically, the calibration process proceeds as follows: First, the corrected geotechnical parameters are input into real-time monitoring data. The corresponding stress response (e.g., strain, displacement, etc.) is calculated based on the current monitoring data. This is then compared with the actual measured values to determine the error. Common error evaluation metrics such as mean square error (RMSE) or mean absolute error (RMSE) can be used to calculate the error. Based on these error values, the system adjusts the correction parameters to reduce the discrepancy between the model's predicted and actual values.

[0059] Specifically, if there is a significant gap between the predicted results of the correction parameters and the subsequent real-time monitoring data, the system will adjust the correction parameters through the feedback mechanism. The formula of the feedback process can be expressed as: E adjusted =E optimized +ΔE; ν adjusted =ν optimized +Δν; Among them, E adjusted is the adjusted elastic modulus of rock and soil, ν adjusted is the adjusted Poisson's ratio, E model and ν modelThe initial elastic modulus and Poisson's ratio based on the model are represented by ΔE, the correction value for the elastic modulus, and Δν, the correction value for the Poisson's ratio. By adjusting ΔE and Δν, the deviation between the predicted and actual values can be reduced, thereby continuously improving the accuracy of the geotechnical parameter correction.

[0060] As an option, an adaptive gain adjustment mechanism can be incorporated into the feedback calibration process to further improve calibration accuracy. Specifically, the feedback calibration adjustment amplitude is controlled by the gain coefficient. The gain coefficient can be adjusted dynamically, and the adjustment step size is optimized based on the error size of the real-time monitoring data. This mechanism can be implemented using the following formula: ΔE = K·RMSE; Δν = K·RMSE; Where ΔE is the correction for the elastic modulus, Δν is the correction for the Poisson's ratio, K is the feedback gain, and RMSE is the root mean square error, representing the difference between the correction result and the actual measurement. Based on the gain adjustment factor, the correction is dynamically adjusted as the error changes, ensuring that each correction effectively narrows the error range.

[0061] In one possible implementation, if the error shown by the monitoring data exceeds a set threshold (such as a maximum error limit), the system automatically enters a multiple-iteration calibration phase, re-evaluating the revised geotechnical parameters until the error is reduced to an acceptable level. This mechanism ensures that the correction process remains stable and effectively avoids deviations caused by error accumulation.

[0062] In some embodiments, the calibration process is not limited to correcting parameters such as stress and strain. It can also correct other parameters such as displacement and settlement based on the specific needs of the project. The calibration process is dynamic and can be adjusted in real time as the project progresses and the external environment changes, ensuring that parameter corrections are always highly consistent with the actual project requirements.

[0063] Furthermore, the feedback calibration mechanism of the present invention is not only applicable to static correction tasks but also capable of rapidly responding to changes in the rock and soil state during construction. During different construction phases, factors such as rock and soil properties, construction loads, and groundwater levels may change, requiring the system to adjust the correction strategy in real time based on the new working condition data. This dynamic calibration process ensures that the mechanical parameters of the rock and soil are accurately corrected in a variety of complex environments.

[0064] The automatic correction system for geotechnical parameters described below and the automatic correction method for geotechnical parameters described above can refer to each other.

[0065] Please see the attached Figure 2 The present invention also provides a rock and soil parameter automatic correction system, including: Sensor networks are used to collect stress and strain data of rock and soil in real time; The data processing unit is used to pre-process the collected data and correct the mechanical parameters of the rock and soil based on the pre-processed data; the optimization module is used to execute the optimization algorithm to optimize the mechanical parameters of the rock and soil to ensure the accuracy of the correction results; The calibration module is used to adjust the correction results according to real-time feedback to ensure the accuracy of the final correction parameters.

[0066] The system of this embodiment can be used to execute the above method embodiments, and its principles and technical effects are similar, so they will not be repeated here.

[0067] The computer device described below and the automatic correction method for rock and soil parameters described above can correspond to each other.

[0068] Please see the attached Figure 3 The present invention further provides a computer device 40, comprising: a processor 41 and a memory 42, wherein the memory 42 stores a computer program executable by the processor, and when the computer program is executed by the processor, the above method is performed.

[0069] The present invention further provides a storage medium 43 on which a computer program is stored. When the computer program is run by the processor 41 , the above method is executed.

[0070] Among them, the storage medium 43 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0071] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for automatically correcting rock and soil parameters, characterized in that: The following steps are involved: Numerical simulation is used to determine the sensor installation location, and the stress and strain parameters of the rock and soil are collected in real time through the sensor network, and the collected data are transmitted to the data processing center; The collected data is pre-processed by the data processing center, using an outlier detection algorithm to detect and remove abnormal data, and using interpolation to repair missing data; Based on the preprocessed data, applying a mechanical behavior model of the rock and soil body to modify key mechanical parameters of the rock and soil body, wherein the key mechanical parameters of the rock and soil body include elastic modulus and Poisson's ratio; Use optimization algorithms to optimize the key mechanical parameters of rock and soil, combining Lagrange multiplier method and variational method to minimize the correction error and meet the constraints; Feedback calibration is performed based on the modified key mechanical parameters of the rock and soil mass. By comparing with subsequent real-time monitoring data, the error is calculated and the modified key mechanical parameters are adjusted based on the feedback.

2. The method for automatic correction of rock and soil parameters according to claim 1, characterized in that: The steps of the numerical simulation include: Analyze the stress and strain distribution of rock and soil under different working conditions through numerical simulation and identify key areas; Design sensor layout based on the results of numerical simulation to ensure monitoring of key areas; Optimize sensor installation locations based on rock and soil characteristics, deformation patterns, and construction conditions to ensure maximum data acquisition accuracy; optimize sensor layout density to ensure comprehensive monitoring and improve data quality; Combining simulation results with actual construction, sensor positions are adjusted to ensure accurate monitoring.

3. The automatic correction method for rock and soil mass parameters according to claim 1, characterized in that: The step of preprocessing the collected data includes: Obtain the original monitoring data transmitted by the sensor, perform preliminary cleaning, and remove noise and irrelevant data; Use standard deviation methods and machine learning algorithms to identify and remove outliers, and perform interpolation repairs for missing or abnormal data to ensure data integrity and accuracy; Evaluate data quality to ensure that the pre-processed data meets the requirements for subsequent corrections.

4. The method for automatic correction of rock and soil parameters according to claim 1, characterized in that: The step of performing dimensionality reduction processing on the collected data includes: Convert the collected high-dimensional data into tensor form and use tensor decomposition technology to extract the main features of the data; The principal component analysis method is used to reduce the dimension of the data and retain the component with the largest variance in the data, thereby reducing redundant information; Feature selection methods are used to further screen the features that are most relevant to security threat analysis.

5. The method for automatic correction of rock and soil parameters according to claim 1, characterized in that: The optimization algorithm includes: using Lagrange multiplier method and variational method to optimize key mechanical parameters of rock and soil, minimize correction errors and meet engineering constraints; The correction results are further optimized through polynomial regression and least squares method to ensure the accuracy of the correction parameters and provide preliminary correction parameters for subsequent feedback calibration.

6. The method for automatic correction of rock and soil parameters according to claim 1, characterized in that: The feedback calibration step includes: Calculating correction errors and adjusting correction coefficients by comparing with subsequent real-time monitoring data, including displacement data, environmental data, construction progress, and load changes; The correction error is evaluated using the mean square error, and the correction coefficient is dynamically adjusted through the adaptive feedback mechanism. The formula is: ΔP adjusted =ΔP initial +γ×ΔP error ; Where ΔP adjusted is the adjusted rock and soil parameter, ΔP initial is the preliminary correction value output by the optimization algorithm, γ is the dynamic inverse constant ΔP error is the error between the actual measurement value and the optimized result; The correction parameters are updated according to the feedback results, and multiple iterations are performed to ensure the accuracy of the final correction results.

7. The method for automatic correction of rock and soil parameters according to claim 1, characterized in that: The key mechanical parameters of the correction are dynamically adjusted through feedback gain and error threshold, and the formula is: E adjusted =E new +K·(RMSE-threshold); Among them, E adjusted is the adjusted elastic modulus of rock and soil, E new is the newly corrected elastic modulus of rock and soil, K is the feedback gain, threshold is the preset error threshold, and RMSE is the root mean square error.

8. A rock and soil mass parameter automatic correction system, applied to the rock and soil mass parameter automatic correction method according to any one of claims 1 to 7, characterized in that: include: Sensor module, used to collect stress and strain data of rock and soil in real time; A data processing module is used to pre-process the collected data and correct the mechanical parameters of the rock and soil based on the pre-processed data; Optimization module, used to optimize the mechanical parameters of rock and soil through optimization algorithms to ensure the accuracy of the correction results; The calibration module is used to adjust the correction results according to real-time feedback to ensure the accuracy of the final correction parameters.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for automatic correction of rock and soil parameters according to any one of claims 1 to 7 is implemented.

10. Automatic correction storage medium for rock and soil parameters, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the method for automatically correcting rock and soil parameters according to any one of claims 1 to 7 is implemented.

Images