A method for predicting the deformation of surrounding rock in tunnel construction in soft rock areas

Abstract

The present invention discloses a method for predicting surrounding rock deformation in tunnel construction in soft rock areas, belonging to the technical field of surrounding rock deformation prediction, and capable of solving the problem of low accuracy of the existing technology in predicting surrounding rock deformation in soft rock areas. The method includes: S1. According to the surrounding rock monitoring network of the tunnel, adaptive weight terms are respectively added to multiple loss functions in the objective function of the gradient boosting decision tree algorithm to obtain an improved gradient boosting decision tree algorithm; the multiple loss functions correspond one by one to multiple monitoring points in the monitoring network; S2. Using the surrounding rock monitoring data collected by the surrounding rock monitoring network to train the improved gradient boosting decision tree algorithm to obtain a surrounding rock deformation prediction model of the tunnel; S3. Using the surrounding rock deformation prediction model to predict the surrounding rock deformation situation of the tunnel. The present invention is used for predicting surrounding rock deformation.

Description

Technical Field

[0001] The present invention relates to a method for predicting surrounding rock deformation in tunnel construction in soft rock areas, belonging to the technical field of surrounding rock deformation prediction. Background Art

[0002] Tunnel construction in soft rock areas is like carving an art piece in soft sand. Every step needs to be cautious and full of challenges. The geological conditions in soft rock areas are like a complex and changeable maze. The interaction between rock layers, the flow of groundwater, the distribution of in-situ stress and other factors are intertwined, making the deformation of the surrounding rock elusive and full of uncertainties. Existing methods for predicting surrounding rock deformation, such as the estimation method based on empirical formulas and the simulation method based on simplified calculation models, are often like blind men feeling an elephant in soft rock areas, and it is difficult to comprehensively, real-timely and accurately reflect the actual deformation of the surrounding rock. This not only brings great safety risks to construction projects in soft rock areas, but also seriously affects the project progress and cost control of construction projects. Summary of the Invention

[0003] The present invention provides a method for predicting surrounding rock deformation in tunnel construction in soft rock areas, which can solve the problems of insufficient accuracy and real-time performance of existing technologies in predicting surrounding rock deformation in soft rock areas.

[0004] The present invention provides a method for predicting surrounding rock deformation in tunnel construction in soft rock areas, and the method includes:

[0005] S1. According to the surrounding rock monitoring network of the tunnel, adaptively weighted terms are respectively added to multiple loss functions in the objective function of the gradient boosting decision tree algorithm to obtain an improved gradient boosting decision tree algorithm; the multiple loss functions correspond one by one to multiple monitoring points in the monitoring network;

[0006] S2. Using the surrounding rock monitoring data collected by the surrounding rock monitoring network, train the improved gradient boosting decision tree algorithm to obtain a surrounding rock deformation prediction model for the tunnel;

[0007] S3. Using the surrounding rock deformation prediction model to predict the surrounding rock deformation of the tunnel.

[0008] Optionally, the S2 is specifically:

[0009] S21. Divide the surrounding rock monitoring data into a training set and a test set, and use the training set to train the improved gradient boosting decision tree algorithm to obtain a trained improved algorithm;

[0010] S22. Use the test set to adjust multiple adaptively weighted terms in the trained improved algorithm to obtain a surrounding rock deformation prediction model for the tunnel.

[0011] Optionally, S22 is specifically:

[0012] According to the test set, and using the trained improved algorithm to predict the deformation conditions of multiple monitoring points, obtaining the deformation prediction effects of each monitoring point;

[0013] Adjusting multiple adaptive weight terms in the trained improved algorithm according to the deformation prediction effects to obtain the surrounding rock deformation prediction model of the tunnel.

[0014] Optionally, in S21, dividing the surrounding rock monitoring data into a training set and a test set is specifically:

[0015] Extracting the characteristic data of the surrounding rock deformation from the surrounding rock monitoring data;

[0016] Dividing the characteristic data of the surrounding rock deformation into a training set and a test set.

[0017] Optionally, in S21, training the improved gradient boosting decision tree algorithm using the training set is specifically:

[0018] Using the training set and training the improved gradient boosting decision tree algorithm by adjusting the learning rate.

[0019] Optionally, before S1, the method further includes:

[0020] Laying out the surrounding rock monitoring network of the tunnel and collecting the surrounding rock monitoring data using the surrounding rock monitoring network.

[0021] Optionally, laying out the surrounding rock monitoring network of the tunnel is specifically:

[0022] Setting multiple monitoring points at different positions in the axial and circumferential directions of the tunnel, and setting one or several of a displacement sensor, a stress sensor, and a temperature sensor at each monitoring point.

[0023] Optionally, before S2, the method further includes:

[0024] Preprocessing the surrounding rock monitoring data collected by the surrounding rock monitoring network;

[0025] Correspondingly, the surrounding rock monitoring data in S2 is the preprocessed surrounding rock monitoring data.

[0026] Optionally, preprocessing the surrounding rock monitoring data collected by the surrounding rock monitoring network is specifically:

[0027] Performing data cleaning, denoising, format conversion, and standardization processing on the surrounding rock monitoring data collected by the surrounding rock monitoring network.

[0028] The beneficial effects that the present invention can produce include:

[0029] By adding an adaptive weight term to the objective function of the gradient boosting decision tree algorithm, the present invention optimizes and improves the algorithm. The obtained surrounding rock deformation prediction model can re-evaluate the importance of each feature according to the deformation prediction effect of each monitoring point, thereby realizing the dynamic adjustment of the model to balance the prediction accuracy and complexity of the model. This improvement enables the model to have better performance and generalization ability when dealing with complex and variable soft rock tunnel data, which is significantly better than the single or simple prediction models that may be used in the prior art.

[0030] By constructing a high-precision surrounding rock monitoring network, the present invention realizes the all-round and real-time monitoring of the surrounding rock deformation, can capture tiny deformation signals, and ensures the comprehensiveness, real-time and accuracy of the data. This enables the data collection to no longer be limited to local or periodic monitoring, but realizes a continuous and real-time data stream, providing a richer and more refined data basis for subsequent data processing and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the surrounding rock deformation prediction method for tunnel construction in soft rock areas provided by an embodiment of the present invention;

[0032] Figure 2 It is a preprocessing flow chart of the surrounding rock monitoring data provided by an embodiment of the present invention;

[0033] Figure 3 It is a construction flow chart of the surrounding rock deformation prediction model provided by an embodiment of the present invention;

[0034] Figure 4 It is a surrounding rock deformation prediction flow chart provided by an embodiment of the present invention;

[0035] Figure 5 It is a schematic diagram of the structure of the deformation prediction system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The present invention will be described in detail below with reference to the embodiments, but the present invention is not limited to these embodiments.

[0037] An embodiment of the present invention provides a surrounding rock deformation prediction method for tunnel construction in soft rock areas, as Figure 1 shown, the method includes:

[0038] S1. According to the surrounding rock monitoring network of the tunnel, an adaptive weight term is respectively added to multiple loss functions in the objective function of the gradient boosting decision tree algorithm to obtain an improved gradient boosting decision tree algorithm; the multiple loss functions correspond one by one to multiple monitoring points in the monitoring network;

[0039] S2. Use the surrounding rock monitoring data collected by the surrounding rock monitoring network to train the improved gradient boosting decision tree algorithm to obtain a surrounding rock deformation prediction model for the tunnel.

[0040] S3. Use the surrounding rock deformation prediction model to predict the surrounding rock deformation of the tunnel.

[0041] In this embodiment, before S1, the method may further include:

[0042] Deploy a surrounding rock monitoring network for the tunnel and use the surrounding rock monitoring network to collect surrounding rock monitoring data.

[0043] Further, the deployment of the surrounding rock monitoring network for the tunnel may specifically be:

[0044] Set a plurality of monitoring points at different positions in the axial and circumferential directions of the tunnel, and set one or several of a displacement sensor, a stress sensor, and a temperature sensor at each monitoring point.

[0045] Specifically, the displacement sensor may be a laser rangefinder, etc., the stress sensor includes a strain gauge or a fiber Bragg grating sensor, etc., and the temperature sensor may be a thermocouple or an infrared thermometer, etc. These sensors are arranged at the monitoring points on the top of the tunnel, the side walls, and other key areas that may be affected by construction to ensure that the deformation dynamics of the surrounding rock can be captured comprehensively and in real time. When specifically arranging, the geometric shape of the tunnel, the geological conditions, and the construction progress should be considered, and a three-dimensional space coordinate system should be used for precise positioning to ensure that the surrounding rock monitoring network can accurately reflect the actual deformation of the surrounding rock.

[0046] Exemplarily, assume that a certain tunnel is located in a soft rock area with complex geological conditions, and the tunnel section is a circle with a diameter of 10 meters. In order to achieve real-time monitoring and accurate prediction of the surrounding rock deformation, a comprehensive surrounding rock monitoring network needs to be arranged at the key positions around the tunnel.

[0047] Specifically, a high-precision laser rangefinder is arranged at the monitoring point at the center position of the tunnel top to monitor the subsidence of the vault. The sensor is 1 meter away from the vault and can capture the minute displacement changes of the vault in real time. Considering the symmetry of the tunnel and the uncertainty of the geological conditions, three displacement sensors are respectively arranged at the upper, middle, and lower monitoring points on the tunnel side walls. The heights of the three monitoring points are 1 meter, 3 meters, and 5 meters away from the tunnel floor respectively, with uniform intervals to ensure comprehensive monitoring of the side wall deformation.

[0048] A displacement sensor is arranged at the monitoring point at the central axis position of the tunnel floor to monitor the heave of the floor. At the same time, displacement sensors are also arranged at the monitoring points 3 meters away from the central axis on both sides of the central axis to monitor the lateral displacement of the floor. The arrangement of these sensors helps to monitor the stability of the tunnel floor and give early warnings for possible floor deformations.

[0049] To monitor the stress conditions of the key support structures in the tunnel, a monitoring point is set every 5 meters near positions such as bolts and shotcrete layers, and stress sensors are arranged at these monitoring points. These stress sensors can withstand harsh underground environments and are very sensitive to stress changes, providing important data support for the stability monitoring of the tunnel. In addition, to consider the influence of environmental temperature on the stability of the surrounding rock, a temperature sensor can also be arranged at each of the five monitoring points at different positions in the tunnel. These sensors can monitor the changes in environmental temperature in real time and respond to the changes in the physical properties of the surrounding rock caused by temperature.

[0050] When arranging the monitoring points, a three-dimensional space coordinate system is adopted for precise positioning to ensure that each monitoring point can be accurately reflected in the three-dimensional model of the tunnel. Through Geographic Information System (GIS) technology, the monitoring points are combined with the geological model of the tunnel to facilitate the analysis and interpretation of the surrounding rock monitoring data. In terms of data collection, all sensors are connected to a central data collection system by wired or wireless means. This system can receive the data from the sensors in real time and transmit it to the data processing center for analysis. The data processing center is equipped with powerful servers and professional data analysis software, which can quickly process and analyze a large amount of data and provide accurate input for the surrounding rock deformation prediction model.

[0051] Using the surrounding rock monitoring network of the tunnel, the surrounding rock monitoring data can be collected to conduct all-round real-time monitoring of the deformation of the surrounding rock during the construction of the tunnel in soft rock areas. This not only improves the safety of tunnel construction but also provides timely deformation warnings and decision-making support for engineering technicians, thereby optimizing the construction process, improving construction efficiency, and reducing construction costs.

[0052] Then, according to the surrounding rock monitoring network of the tunnel, this embodiment improves the Gradient Boosting Decision Tree (GBDT) algorithm to obtain an improved Gradient Boosting Decision Tree algorithm (improved GBDT algorithm).

[0053] The improved GBDT algorithm introduces a feature importance re-evaluation strategy on the basis of the GBDT algorithm. Among them, the feature is a key index that has an important impact on the deformation of the surrounding rock, including historical deformation amount, deformation rate, temperature change, stress state, etc. The feature importance re-evaluation strategy iteratively re-evaluates the importance of each feature and adjusts the selection and weight of each feature accordingly to ensure that the finally established surrounding rock deformation prediction model can capture the most critical information for the prediction result.

[0054] Specifically, in this embodiment, the objective function of the GBDT algorithm is optimized and improved. The objective function of the GBDT algorithm generally includes two parts: a loss function and a regularization term. In this embodiment, an adaptive weight term is added to each loss function in the objective function of the GBDT algorithm to improve this objective function, enabling it to better balance the prediction accuracy and complexity of the model.

[0055] The objective function of the improved GBDT algorithm is expressed as:

[0056] (1)

[0057] In formula (1), is the objective function of the improved GBDT algorithm; is the loss function of monitoring point , which is used to measure the difference between the predicted deformation amount and the true deformation amount of monitoring point . The smaller the difference, the more accurate the prediction of the model; is the true deformation amount of monitoring point ; is the predicted deformation amount of monitoring point ; is the feature that has an impact on the predicted deformation amount of monitoring point ; is the total number of monitoring points; is the adaptive weight term of monitoring point , which is dynamically adjusted according to the prediction difficulty of monitoring point ; is the regularization term, which is used to control the complexity of the surrounding rock deformation prediction model; and are regularization parameters; is the total number of features; is the importance weight of feature ; is an indicator function. When feature is selected, takes a value of 1, otherwise takes a value of 0.

[0058] Specifically, for regression problems, common loss functions include Mean Square Error (MSE for short); for classification problems, common loss functions include cross-entropy loss.

[0059] By introducing the adaptive weight term , the surrounding rock deformation prediction model can pay more attention to those monitoring points with greater prediction difficulty. Before the start of model training, this embodiment assigns an adaptive weight term , its initial value can be set according to the category, distribution location, prediction difficulty, prediction requirements, etc. of the monitoring points. For example, when dealing with an imbalanced dataset, a higher weight can be assigned to the monitoring points corresponding to the minority class samples.

[0060] During the model training process, the adaptive weight term is dynamically adjusted according to the prediction performance of the model at each monitoring point. If the prediction error at a certain monitoring point is large, its weight value can be increased to make the model pay more attention to these monitoring points in subsequent iterations. This dynamic adjustment mechanism helps to improve the prediction accuracy of the model for monitoring points with greater prediction difficulty, thereby improving the overall prediction performance of the model.

[0061] Specifically, common regularization terms include L1 regularization (the sum of the absolute values of the weights) and L2 regularization (the sum of the squares of the weights).

[0062] Regularization term can control the complexity of the model and prevent the model from overfitting. Overfitting means that the model performs well on the training data but poorly on new, unseen data. Regularization improves the generalization ability of the model by restricting its complexity. Among them, and need to determine their optimal values through methods such as cross-validation to balance the fitting ability and generalization ability of the model.

[0063] In summary, the objective function of the improved GBDT algorithm can better balance the prediction accuracy and complexity of the model, and improve the prediction performance and generalization ability of the model in the deformation prediction task of tunnel construction in soft rock areas.

[0064] In this embodiment, surrounding rock monitoring data is collected through a surrounding rock monitoring network, and the improved GBDT algorithm is trained using the surrounding rock monitoring data to obtain a surrounding rock deformation prediction model for the tunnel.

[0065] Before training, that is, before S2, this embodiment also needs to preprocess the surrounding rock monitoring data collected by the surrounding rock monitoring network;

[0066] Correspondingly, the surrounding rock monitoring data in S2 is the preprocessed surrounding rock monitoring data.

[0067] Furthermore, the preprocessing of the surrounding rock monitoring data collected by the surrounding rock monitoring network can specifically be:

[0068] Perform data cleaning, denoising, format conversion, and standardization on the surrounding rock monitoring data collected by the surrounding rock monitoring network.

[0069] Specifically, the data preprocessing process of this embodiment is as Figure 2As shown. The data processing center receives the surrounding rock monitoring data collected by the surrounding rock monitoring network. First, it cleans the surrounding rock monitoring data to remove the outliers or missing values caused by equipment failures, signal interferences, etc. Then, it uses digital filtering techniques (such as Kalman filtering) to denoise the cleaned data to reduce the impact of random errors on subsequent analyses. Finally, it performs format conversion and standardization processing on the denoised data to ensure that the data collected by different sensors can be unified, facilitating subsequent data processing and model training.

[0070] Specifically, the recurrence formula of Kalman filtering includes a prediction equation and an update equation.

[0071] Among them, the prediction equation can be expressed as:

[0072] (2)

[0073] (3)

[0074] In equations (2) and (3), is the prediction of the state at time based on the estimate at time. "State" refers to the current condition of the system, such as displacement, velocity, acceleration, etc. For example; represents the displacement vector at represents the covariance matrix at is the state transition matrix; is the control matrix; is the control input; is the prediction error covariance matrix; is the process noise covariance matrix.

[0075] The update equation can be expressed as:

[0076] (4)

[0077] (5)

[0078] (6)

[0079] In equations (4) to (6), is the Kalman gain; is the observation matrix; is the observation value at is the observation noise covariance matrix; is the updated state estimate; is the updated error covariance matrix.

[0080] In this embodiment, S2 can specifically be:

[0081] S21. Divide the preprocessed surrounding rock monitoring data into a training set and a test set, and use the training set to train the improved gradient boosting decision tree algorithm to obtain the trained improved algorithm;

[0082] S22. Use the test set to adjust multiple adaptive weight terms in the trained improved algorithm to obtain a prediction model for the deformation of the surrounding rock of the tunnel.

[0083] Furthermore, in S21, dividing the preprocessed surrounding rock monitoring data into a training set and a test set can specifically be:

[0084] Extract the characteristic data of the surrounding rock deformation from the preprocessed surrounding rock monitoring data; divide the characteristic data of the surrounding rock deformation into a training set and a test set.

[0085] Specifically, in this embodiment, through feature engineering, the characteristic data that has an important impact on the surrounding rock deformation is extracted from the preprocessed surrounding rock monitoring data, such as historical deformation data, deformation rate data, temperature change data, stress state data, etc., to form a data set. Then, the data set is divided into a training set and a test set by using techniques such as cross-validation, which are used for the training and verification of the surrounding rock deformation prediction model respectively.

[0086] Furthermore, in S21, using the training set to train the improved gradient boosting decision tree algorithm to obtain the trained improved algorithm can specifically be:

[0087] Use the training set and train the improved gradient boosting decision tree algorithm by adjusting the learning rate to obtain the trained improved algorithm.

[0088] In this way, the improved gradient boosting decision tree algorithm can dynamically adjust the learning rate according to the descent speed of the loss function during the training process, thereby accelerating the convergence speed and reducing the risk of overfitting.

[0089] Specifically, in this embodiment, the learning rate is dynamically adjusted according to the training process to optimize the model performance. Adjusting the learning rate includes:

[0090] Gradient accumulation: In each iteration, accumulate the squares of all past gradients for adjusting the learning rate. For example, in the AdaGrad algorithm, the learning rate is adjusted using the square root inverse proportion.

[0091] Decay coefficient: In the RMSprop algorithm, a decay coefficient is introduced to prevent the accumulation of the sum of gradient squares from being too extreme, thereby avoiding a sharp decrease in the learning rate.

[0092] Momentum and variance estimation: In the Adam algorithm, by combining the advantages of RMSprop and Momentum, the first-order moment estimation and second-order moment estimation of the gradient are used to adjust the learning rate.

[0093] Furthermore, S22 can specifically be:

[0094] According to the test set, and using the trained improved algorithm to predict the deformation conditions of multiple monitoring points, the deformation prediction effects of each monitoring point are obtained;

[0095] According to the deformation prediction effects, adjust multiple adaptive weight terms in the trained improved algorithm to obtain a surrounding rock deformation prediction model for the tunnel.

[0096] Specifically, in this embodiment, a testing machine is used to verify the prediction effects of the trained improved algorithm at each monitoring point, and the adaptive weight terms and other parameters are dynamically adjusted according to the prediction effects to optimize the algorithm parameters and the model structure. If the prediction error of a certain monitoring point is large, its weight value can be increased, so that the finally established surrounding rock deformation prediction model pays more attention to these monitoring points in subsequent iterations. This dynamic adjustment mechanism helps to improve the prediction accuracy of the surrounding rock deformation prediction model for monitoring points with greater prediction difficulty, thereby improving the prediction performance of the surrounding rock deformation prediction model. The model construction process of this embodiment is as Figure 3 shown.

[0097] After the surrounding rock deformation prediction model is constructed, in this embodiment, the real-time collected surrounding rock monitoring data is input into the prediction model for surrounding rock deformation prediction. The prediction model will calculate and output key indicators such as the deformation amount and deformation rate of the surrounding rock based on the input data, which is used to guide the construction. To ensure the real-time nature of the prediction, this embodiment uses a stream processing framework (such as Apache Kafka or Spark Streaming) to process and analyze the real-time data stream.

[0098] In summary, the flow chart of the surrounding rock deformation prediction in this embodiment is as Figure 4 shown.

[0099] In terms of data collection, the data collection and processing algorithm in this embodiment realizes the full-range and real-time monitoring of the surrounding rock deformation through a high-precision surrounding rock monitoring network, can capture tiny deformation signals, and ensures the comprehensiveness, real-time nature, and accuracy of the data. This makes the data collection no longer limited to local or periodic monitoring, but realizes a continuous and real-time data stream, providing a richer and more refined data basis for subsequent data processing and analysis.

[0100] In terms of data processing, this embodiment adopts advanced data cleaning, denoising processing, and feature extraction technologies, such as digital filtering technologies like Kalman filtering, to reduce the impact of random errors on subsequent analysis and ensure that the data collected by different sensors can be in a unified format, facilitating subsequent data processing and model training. This refined data processing process can extract valuable information more effectively compared to the simple data processing methods in the prior art, providing more reliable inputs for the prediction model.

[0101] In terms of constructing the prediction model, the prediction model of this solution is constructed based on big data and machine learning technologies, which can comprehensively consider the impacts of various factors such as geological conditions, construction methods, and support measures on the surrounding rock deformation to achieve accurate prediction. In particular, this embodiment adopts an improved GBDT algorithm, introducing an adaptive learning rate mechanism and a feature importance re-evaluation strategy, and optimizing the objective function of the algorithm, enabling it to better balance the prediction accuracy and complexity of the model. This improvement enables the model to have better performance and generalization ability when dealing with complex and variable soft rock tunnel data, significantly superior to the single or simple prediction models in the prior art.

[0102] In addition, the prediction model of this embodiment also has a strong adaptive ability and a real-time feedback mechanism, which can dynamically adjust and optimize the prediction model according to real-time monitoring data to adapt to the changes in complex and variable geological conditions and construction environments. This adaptive and real-time feedback ability makes the prediction model of this solution not only superior to the prior art in terms of prediction accuracy but also significantly improved in terms of real-time performance and adaptability.

[0103] It can be seen that the improvements in the data acquisition processing algorithm and the prediction model algorithm in this embodiment not only improve the comprehensiveness, real-time performance, and accuracy of data acquisition but also enhance the refinement level of data processing and the accuracy, real-time performance, and adaptability of the prediction model, thus providing a more scientific and effective solution for the prediction of surrounding rock deformation during the construction of tunnels in soft rock areas. These improvements make this embodiment significantly different from the prior art in terms of technical implementation and application effects, providing stronger guarantees for the safety and engineering quality of tunnel projects.

[0104] Another embodiment of the present invention provides a deformation prediction system based on the above-mentioned surrounding rock deformation prediction method for tunnel construction in soft rock areas, as Figure 5 shown.

[0105] The deformation prediction system includes a data acquisition module, a data transmission module, a data processing module, a prediction model module and a result output module. The data acquisition module uses the surrounding rock monitoring network to collect real-time data of the surrounding rock; the data transmission module transmits the collected data to the data processing module through the data transmission system; the data processing module preprocesses and extracts features of the data; the prediction model module predicts deformation based on machine learning algorithms; the result output module visualizes the prediction results, generates deformation warning information and construction guidance information, and outputs them to engineering and technical personnel.

[0106] The deformation prediction system uses advanced sensor technology and data analysis algorithms to achieve real-time monitoring and accurate prediction of surrounding rock deformation. Through the surrounding rock monitoring network arranged around the tunnel, the system can collect surrounding rock deformation data in real time and transmit the data to the data center for processing and analysis through high-speed data transmission technology. The data analysis algorithm will combine geomechanics theory, construction experience and real-time monitoring data to establish an accurate prediction model to accurately predict the deformation trend of the surrounding rock.

[0107] In addition, the deformation prediction system also has strong adaptive capabilities and real-time feedback mechanisms. The system can dynamically adjust and optimize the prediction model based on real-time monitoring data to adapt to complex and changing geological conditions and changes in the construction environment. At the same time, the system can also provide deformation warning information and construction guidance information to engineering and technical personnel in real time, helping them to take effective measures in a timely manner to deal with potential safety risks.

[0108] Therefore, the deformation prediction system not only improves the accuracy and real-time performance of the prediction results, but also enhances the adaptability and practicality of the system. Its emergence will provide a more scientific and effective solution to the deformation prediction problem during tunnel construction in soft rock areas, and promote the continuous progress and development of tunnel engineering technology.

[0109] The above are only a few embodiments of the present application and do not constitute any form of limitation to the present application. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any technician familiar with the profession, without departing from the scope of the technical solution of the present application, using the technical content disclosed above to make slight changes or modifications are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. A method for predicting the deformation of surrounding rock in tunnel construction in soft rock areas, characterized in that, The method includes: S1. According to the surrounding rock monitoring network of the tunnel, adaptively weighted terms are respectively added to multiple loss functions in the objective function of the gradient boosting decision tree algorithm to obtain an improved gradient boosting decision tree algorithm; the multiple loss functions correspond one-to-one to multiple monitoring points in the monitoring network; S2. Using the surrounding rock monitoring data collected by the surrounding rock monitoring network to train the improved gradient boosting decision tree algorithm to obtain a surrounding rock deformation prediction model of the tunnel; S3. Using the surrounding rock deformation prediction model to predict the surrounding rock deformation condition of the tunnel; The expression of the objective function of the improved gradient boosting decision tree algorithm is: ， In the formula, is the objective function of the improved gradient boosting decision tree algorithm, is the loss function of the monitoring point ; is the true deformation of the monitoring point ; is the predicted deformation of the monitoring point ; is the feature that affects the predicted deformation of the monitoring point ; is the total number of monitoring points, is the adaptive weight term of the loss function of the monitoring point ; is the regularization term, and are the regularization parameters, is the total number of features, is the importance weight of the feature ; is the indicator function; The specific content of S2 is: S21. Divide the surrounding rock monitoring data into a training set and a test set, and use the training set to train the improved gradient boosting decision tree algorithm to obtain a trained improved algorithm; S22. Use the test set to adjust multiple adaptively weighted terms in the trained improved algorithm to obtain a surrounding rock deformation prediction model of the tunnel.

2. The method according to claim 1, wherein The specific content of S22 is: According to the test set, and use the trained improved algorithm to predict the deformation conditions of multiple monitoring points to obtain the deformation prediction effect of each monitoring point; Adjust multiple adaptively weighted terms in the trained improved algorithm according to the deformation prediction effect to obtain a surrounding rock deformation prediction model of the tunnel.

3. The method according to claim 1, wherein In S21, dividing the surrounding rock monitoring data into a training set and a test set specifically includes: Extract the characteristic data of the surrounding rock deformation from the surrounding rock monitoring data; Divide the characteristic data of the surrounding rock deformation into a training set and a test set.

4. The method according to claim 1, characterized in that, In S21, using the training set to train the improved gradient boosting decision tree algorithm specifically includes: Using the training set, and training the improved gradient boosting decision tree algorithm by adjusting the learning rate.

5. The method according to claim 1, wherein Before S1, the method further includes: Deploy the surrounding rock monitoring network of the tunnel, and use the surrounding rock monitoring network to collect the surrounding rock monitoring data.

6. The method according to claim 5, characterized in that, The specific content of deploying the surrounding rock monitoring network of the tunnel is: Set multiple monitoring points at different positions in the axial and circumferential directions of the tunnel, and set one or several of a displacement sensor, a stress sensor, and a temperature sensor at each monitoring point.

7. The method according to claim 1, characterized in that, Before S2, the method further includes: Preprocess the surrounding rock monitoring data collected by the surrounding rock monitoring network; Correspondingly, the surrounding rock monitoring data in S2 is the preprocessed surrounding rock monitoring data.

8. The method according to claim 7, wherein The specific content of preprocessing the surrounding rock monitoring data collected by the surrounding rock monitoring network is: Perform data cleaning, noise reduction processing, format conversion, and standardization processing on the surrounding rock monitoring data collected by the surrounding rock monitoring network.

Images