Monitoring and early warning system for geotechnical engineering

Abstract

The present invention discloses a monitoring and early warning system for geotechnical engineering, which relates to the technical field of geological disaster early warning. The monitoring and early warning system for geotechnical engineering includes a data acquisition module, a data processing module, a model building module and a disaster early warning module. The present invention predicts the displacement and shear stress of the rock and soil body through a displacement-hydrological-geological coupling prediction model, which can not only more comprehensively reflect the mechanical state of the rock and soil body and the comprehensive impact of environmental factors on geotechnical engineering disasters, but also comprehensively consider the coupling effect of various monitoring data, thereby effectively reducing the prediction error and uncertainty caused by data missing or a single model. By processing the hydrological data through causal convolution, it can not only effectively capture the long-term dependence between the hydrological data, but also reflect the correlation and hysteresis between the water level, rainfall and pore water pressure, thereby significantly improving the accuracy of the model prediction and also improving the early warning accuracy of this monitoring and early warning system.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster early warning technology, and in particular to a monitoring and early warning system for geotechnical engineering. Background Art

[0002] Geotechnical engineering, a branch of civil engineering, focuses on the behavioral characteristics of natural or artificially modified rock and soil materials and their applications in engineering construction. It encompasses a wide range of fields, from geological surveys, foundation treatment, slope stability analysis, to tunnel and underground space development. Due to the heterogeneity, anisotropy, and time-dependent characteristics of rock and soil, their behavior is often complex and variable, making them prone to geotechnical engineering hazards. These hazards are of varying types.

[0003] Existing monitoring and early warning systems for geotechnical engineering mainly monitor key geotechnical parameters such as displacement, stress, and seepage pressure in real time through sensors and other equipment, and then predict the occurrence of geotechnical engineering disasters through machine learning models and take corresponding measures. However, this early warning method only considers the relationship between monitoring data and geotechnical engineering disasters such as landslides, ignoring the correlation and coupling effects between monitoring data. For example, the triggering of a landslide may be caused by a chain reaction of "rainfall infiltration causing pore water pressure to rise, which then reduces the shear strength of the geotechnical soil and finally leads to a sudden change in displacement", resulting in poor prediction accuracy. In addition, when predicting and outputting results, existing machine learning models may learn incorrect rules due to noise and bias in the input data, resulting in output results that do not conform to reality and physical laws, thus affecting the rationality of the prediction.

[0004] Based on the above situation, the present invention proposes a monitoring and early warning system for geotechnical engineering with accurate and reasonable early warning. Summary of the Invention

[0005] In order to overcome the shortcomings of existing monitoring and early warning systems for geotechnical engineering, which only consider the relationship between monitoring data and geotechnical engineering disasters such as landslides, and ignore the correlation and coupling effect between monitoring data, resulting in poor prediction accuracy, and the possibility that when outputting prediction results, the machine learning model may learn wrong rules due to noise and deviation in the input data, and the output results may not conform to the actual situation and physical laws, thereby affecting the rationality of the prediction, the present invention proposes a monitoring and early warning system for geotechnical engineering with accurate and reasonable warning.

[0006] Monitoring and early warning systems for geotechnical engineering include:

[0007] The data acquisition module is used to obtain the historical displacement data and historical hydrological data of the rock and soil body to be predicted through sensors, and obtain the historical geological data of the rock and soil body through geological exploration reports, wherein the historical geological data includes static geological data and dynamic geological data;

[0008] A data processing module is used to pre-process the acquired historical displacement data, historical hydrological data and dynamic geological data to obtain pre-processed historical displacement data, historical hydrological data and dynamic geological data;

[0009] The model building module is used to construct a displacement-hydrology-geology coupled prediction model. During model training, the Mohr-Coulomb criterion is incorporated into the model's loss function as a regularization term. The prediction error of the model is calculated using the loss function, and the model is iteratively optimized through backpropagation and parameter updates.

[0010] The disaster warning module is used to input the static geological data of the rock and soil to be predicted, the pre-processed historical displacement data, the historical hydrological data and the dynamic geological data into the trained displacement-hydrological-geological joint prediction model to obtain the prediction results, which include the displacement curve of the rock and soil in the next 72 hours. and shear stress curves , based on the prediction results, the rock and soil mass is divided into different risk levels and the corresponding level of alarm is activated.

[0011] As a preferred aspect of the invention, the historical displacement data is the displacement of the rock and soil body. Time series, the historical hydrological data include instantaneous rainfall intensity time series, groundwater level height time series and pore water pressure time series, the static geological data include rock and soil type, slope, vegetation coverage, crack development index, cohesion and internal friction angle The dynamic geological data is the normal stress of the rock and soil mass Time series and shear stress Time series.

[0012] As a preferred aspect of the invention, the displacement-hydrology-geology coupling prediction model includes three input layers, three feature extraction layers, a spatiotemporal fusion layer and two joint prediction layers, wherein the three input layers are: a time series input layer for receiving historical displacement data and dynamic geological data; a hydrological sequence input layer for receiving historical hydrological data; a geological parameter input layer for receiving static geological data; wherein the three feature extraction layers are: an LSTM encoder layer for extracting features from the time series received by the time series input layer; a temporal convolutional network layer for extracting local features from the time series received by the hydrological sequence input layer; a fully connected embedding layer for mapping the high-dimensional geological information received by the geological parameter input layer into a low-dimensional feature vector; wherein the spatiotemporal fusion layer is used to splice the outputs of the three feature extraction layers into a feature vector, and weightedly fuse the spliced ​​feature vectors through the attention mechanism to obtain the final fused feature vector; wherein the two joint prediction layers are: a displacement prediction layer for outputting the displacement curve for the next 72 hours ; Shear stress prediction layer, used to output the shear stress curve for the next 72 hours .

[0013] As a preferred aspect of the invention, the training steps of the displacement-hydrology-geology coupled prediction model are specifically as follows:

[0014] Data preparation and preprocessing: obtaining historical displacement data, historical hydrological data, and historical geological data of the sample rock and soil mass, preprocessing all the acquired data, and dividing all the preprocessed data into training sets, validation sets, and test sets in proportion;

[0015] Forward propagation and loss calculation, input the training set into the displacement-hydrology-geology coupling prediction model, and obtain the predicted displacement curve and predicted shear stress curves , according to the predicted displacement curve and actual displacement curve The difference between the two gives the displacement loss term , according to the predicted shear stress curve and ultimate shear stress curve The difference between them gives the physical law loss term based on the Mohr-Coulomb criterion , the total loss value is obtained by calculating , the calculation formula is: ,in is the regularization strength coefficient;

[0016] Back propagation and parameter update, from the total loss value through the chain rule Reversely calculate the gradient of each layer's parameters and adjust the parameters using the gradient descent algorithm;

[0017] Iterative optimization: divide the training set into multiple small batches, repeat the previous two steps batch by batch, and traverse the entire training set multiple times until the model converges or reaches the preset number of iterations;

[0018] Verification and testing: Periodically evaluate the model's losses and indicators through the validation set to prevent overfitting. After training is completed, the performance of the prediction model is evaluated through the test set.

[0019] As a preferred aspect of the invention, the displacement loss term is obtained The specific steps are:

[0020] According to the displacement after preprocessing Time series to obtain the actual displacement curve ;

[0021] The displacement loss term is obtained by calculation , the calculation formula is: ,in Represents the displacement after preprocessing The number of data in the time series, Represents the displacement after preprocessing The first The actual displacement value, Indicates the The predicted displacement value corresponding to the actual displacement value.

[0022] As a preferred aspect of the invention, the physical law loss term is obtained The specific steps are:

[0023] According to cohesion , internal friction angle and normal stress after pretreatment Time series and ultimate shear stress curve obtained by calculation , the calculation formula is: ;

[0024] The physical law loss term based on the Mohr-Coulomb criterion is obtained by calculation , the calculation formula is: ,in represents the normal stress after pretreatment The number of data in the time series, Indicates the Normal stress The predicted shear stress corresponding to the data, Indicates the Normal stress The ultimate shear stress corresponding to the data is, Indicates taking and 0, whichever is greater.

[0025] As a preferred aspect of the present invention, the specific steps of classifying the rock and soil mass into different risk levels according to the prediction results and activating the corresponding level alarms are:

[0026] According to the displacement curve of the rock and soil in the next 72 hours Obtain the displacement velocity curve of the rock and soil ;

[0027] According to the displacement velocity curve Obtain the maximum displacement velocity of the rock mass within the next 72 hours , based on the preset warning level classification rules and maximum displacement speed The rock and soil bodies are classified into risk levels, and the corresponding level of alarm is activated according to the classified risk level and corresponding response measures are recommended.

[0028] The present invention has the following advantages:

[0029] 1. The present invention predicts the displacement and shear stress of rock and soil through a displacement-hydrology-geology coupling prediction model, which can not only more comprehensively reflect the mechanical state of rock and soil and the comprehensive influence of environmental factors on geotechnical engineering disasters, but also comprehensively consider the coupling effect of various monitoring data, thereby effectively reducing the prediction error and uncertainty caused by data missing or single model, so as to significantly improve the accuracy of disaster prediction. The causal convolution in the temporal convolutional network is used to process hydrological data, which can not only effectively capture the long-term dependence between hydrological data, but also reflect the correlation and hysteresis between water level, rainfall and pore water pressure, thereby significantly improving the accuracy of model prediction and the warning accuracy of this monitoring and early warning system.

[0030] 2. The present invention incorporates the Mohr-Coulomb criterion as a regularization term into the loss function of the model, which can physically constrain the output of the displacement-hydrology-geology coupling prediction model. This not only directly avoids the result of the model output conflicting with the principles of geotechnical mechanics, thereby ensuring that the model output conforms to physical laws and can be explained by mechanical theory, thereby improving the physical credibility and interpretability of the prediction results, but also suppresses the model's sensitivity to noise or outliers in the training data, and balances the advantages of data-driven and physical laws, thereby effectively preventing complex models from overfitting and significantly improving the physical rationality and generalization ability of the model, thereby improving the warning rationality of this monitoring and early warning system.

[0031] 3. The present invention divides the risk levels of rock and soil bodies into different levels through early warning level classification rules, and activates the corresponding level of alarm based on the classified risk levels, which can realize graded early warning, thereby not only reminding relevant personnel to take corresponding preventive measures in advance and reasonably allocate limited resources to improve the efficiency of early warning and response, avoid waste of resources, and effectively avoid or reduce disaster losses, but also provide decision makers with clear decision-making basis, so that they can reasonably arrange monitoring, early warning and emergency response and other tasks according to the current risk situation, optimize resource allocation, improve management efficiency, so as to ensure the scientificity and effectiveness of disaster prevention and mitigation work, and enhance the practicality of this monitoring and early warning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a schematic diagram of the structure of a monitoring and early warning system for geotechnical engineering adopted in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0034] Example 1, a monitoring and early warning system for geotechnical engineering, such as Figure 1 Shown, including:

[0035] The data acquisition module is used to obtain the historical displacement data and historical hydrological data of the rock and soil body to be predicted through sensors, and obtain the historical geological data of the rock and soil body through geological exploration reports, wherein the historical geological data includes static geological data and dynamic geological data;

[0036] A data processing module is used to pre-process the acquired historical displacement data, historical hydrological data and dynamic geological data to obtain pre-processed historical displacement data, historical hydrological data and dynamic geological data;

[0037] The model building module is used to construct a displacement-hydrology-geology coupled prediction model. During model training, the Mohr-Coulomb criterion is incorporated into the loss function of the displacement-hydrology-geology coupled prediction model as a regularization term to obtain a loss function with physical constraints. The prediction error of the model is calculated using the loss function with physical constraints, and the model is iteratively optimized through backpropagation and parameter updates.

[0038] The disaster warning module is used to input the static geological data of the rock and soil to be predicted, the pre-processed historical displacement data, the historical hydrological data and the dynamic geological data into the trained displacement-hydrological-geological joint prediction model to obtain the prediction results, which include the displacement curve of the rock and soil in the next 72 hours. and shear stress curves , based on the prediction results of the displacement-hydrology-geology joint prediction model, the rock and soil mass is divided into different risk levels, and the corresponding level of alarm is activated.

[0039] The historical displacement data is composed of the historical millimeter-level displacement changes of the rock and soil body. The historical hydrological data include instantaneous rainfall intensity time series, groundwater level height time series and pore water pressure time series, and the static geological data include rock and soil type, slope, vegetation coverage, crack development index, cohesion and internal friction angle , and the dynamic geological data is the normal stress of the rock and soil Time series and shear stress Time series.

[0040] The specific steps of pre-processing the acquired historical displacement data, historical hydrological data and dynamic geological data are as follows:

[0041] Time alignment: unify the starting base time point of all time series, unify the time frequency of all time series, and fill the missing values ​​of time series with low time frequency at the new time point through linear interpolation or polynomial interpolation to ensure that all time series data have values ​​at the same time point;

[0042] Missing value processing: Processing of null or missing values ​​in time series, using linear interpolation or polynomial interpolation to fill missing values, or directly using the mean or median of the time series to fill missing values, to ensure data integrity and model stability;

[0043] Outlier processing: using the Z-Score method or the IQR method to identify outliers in the time series that do not conform to the expected pattern, and then removing outliers and replacing them with the mean or median of the time series or using interpolation methods to repair outliers to avoid these outliers from negatively affecting model training;

[0044] Data standardization: standardize data of different dimensions through minimum-maximum standardization. The calculation formula of minimum-maximum standardization is: ,in Represents the original data value, represents the minimum value in the time series, represents the maximum value in the time series, Represents the normalized data value.

[0045] The displacement-hydrology-geology coupling prediction model includes three input layers, three feature extraction layers, a spatiotemporal fusion layer and two joint prediction layers, wherein the three input layers are: a time series input layer for receiving displacement Time series, normal stress Time series and shear stress Time series; hydrological sequence input layer, used to receive instantaneous rainfall intensity time series, groundwater level height time series and pore water pressure time series; geological parameter input layer, used to receive static geological data; the three feature extraction layers are: LSTM encoder layer, the structure is a bidirectional LSTM network fused with a self-attention mechanism, used to extract features of the time series received by the time series input layer through a bidirectional LSTM model fused with a self-attention mechanism, and output a 128-dimensional feature vector; temporal convolutional network layer, used to extract local features of the time series received by the hydrological sequence input layer through the void convolution and causal convolution in the temporal convolutional network, and output 64-dimensional feature matrix; fully connected embedding layer, the structure is a three-layer fully connected network, which is used to map the high-dimensional geological information received by the geological parameter input layer into a low-dimensional feature vector through the three-layer fully connected network, and output a 64-dimensional feature vector; the spatiotemporal fusion layer is used to splice the outputs of the three feature extraction layers into a feature vector, and perform weighted fusion of the spliced ​​feature vectors through the attention mechanism to obtain the final fused feature vector; the two joint prediction layers are: displacement prediction layer, the structure is a 3-layer TCN network + 1-layer linear regression layer, which is used to extract time series features from the fused feature vector through the time convolution network, and output the displacement curve for the next 72 hours through the linear regression layer. The shear stress prediction layer consists of a 3-layer TCN network and a 1-layer linear regression layer. It is used to extract the time series features of the fused feature vector through the time convolution network and output the shear stress curve for the next 72 hours through the linear regression layer. .

[0046] It should be noted that the above steps predict the displacement and shear stress of the rock and soil body through the displacement-hydrology-geology coupling prediction model, which can not only more comprehensively reflect the mechanical state of the rock and soil body and the comprehensive influence of environmental factors on geotechnical engineering disasters, but also comprehensively consider the coupling effect of various monitoring data, thereby effectively reducing the prediction error and uncertainty caused by data missing or single model, so as to significantly improve the accuracy of disaster prediction. The processing of hydrological data through causal convolution in the temporal convolutional network can not only effectively capture the long-term dependence between hydrological data, but also reflect the correlation and hysteresis between water level, rainfall and pore water pressure, thereby significantly improving the accuracy of model prediction and the warning accuracy of this monitoring and early warning system.

[0047] The training steps of the displacement-hydrology-geology coupled prediction model are specifically as follows:

[0048] Data preparation and preprocessing: historical displacement data, historical hydrological data, and historical geological data of the sample rock and soil are obtained. All the above-obtained data are time-aligned, missing value processing, outlier processing, and data standardization are performed. All preprocessed data are divided into training set, validation set, and test set in the ratio of 70%, 15%, and 15%, respectively;

[0049] Forward propagation and loss calculation, the sample data in the training set is input into the displacement-hydrology-geology coupling prediction model, and the predicted displacement curve is finally obtained through linear transformation and nonlinear activation of each layer. and predicted shear stress curves , according to the predicted displacement curve and actual displacement curve The difference between the two gives the displacement loss term , according to the predicted shear stress curve and ultimate shear stress curve The difference between them gives the physical law loss term based on the Mohr-Coulomb criterion , the total loss value is calculated by the total loss calculation formula , the calculation formula is: ,in is the regularization strength coefficient, which is used to control the physical law loss term The weight of , and it is necessary to use grid search optimization and other parameter adjustment strategies to Adjust the value of ;

[0050] Back propagation and parameter update, from the total loss value through the chain rule Back-calculate the gradient of each layer parameter and use gradient descent algorithms such as SGD and Adam to adjust the parameters to minimize the total loss value ;

[0051] Iterative optimization: divide the training set into multiple small batches, repeat the previous two steps batch by batch, and traverse the entire training set multiple times until the model converges or reaches the preset number of iterations;

[0052] Verification and testing: The losses and indicators of the displacement-hydrology-geology coupling prediction model are periodically evaluated through sample data in the validation set to prevent overfitting. After training is completed, the final performance of this prediction model is evaluated through sample data in the test set.

[0053] The predicted displacement curve and actual displacement curve The difference between the two gives the displacement loss term The specific steps are:

[0054] According to the preprocessed displacement in the historical displacement data Time series to obtain the actual displacement curve ;

[0055] The displacement loss term is calculated by the mean square error formula , the calculation formula is: ,in Represents the displacement after preprocessing The number of data in the time series, Represents the displacement after preprocessing The first The actual displacement value, Indicates the The predicted displacement value corresponding to the actual displacement value.

[0056] The predicted shear stress curve and ultimate shear stress curve The difference between them gives the physical law loss term based on the Mohr-Coulomb criterion The steps are:

[0057] According to the cohesion in static geological data , internal friction angle and preprocessed normal stress in dynamic geological data Time series and ultimate shear stress curve obtained by calculation , the calculation formula is: ;

[0058] The physical law loss term based on the Mohr-Coulomb criterion is calculated by the calculation formula of the regularized loss term , the calculation formula is: ,in represents the normal stress after pretreatment The number of data in the time series, Indicates the Normal stress The predicted shear stress corresponding to the data, Indicates the Normal stress The ultimate shear stress corresponding to the data is, Indicates taking and 0, that is, if , then take 0.

[0059] It should be noted that the above steps incorporate the Mohr-Coulomb criterion as a regularization term into the loss function of the model, which can physically constrain the output of the displacement-hydrology-geology coupling prediction model. This not only directly avoids the conflict between the model output and the principles of geotechnical mechanics, thereby ensuring that the model output conforms to physical laws and can be explained by mechanical theory, thereby improving the physical credibility and interpretability of the prediction results, but also suppresses the model's sensitivity to noise or outliers in the training data, and balances the advantages of data-driven and physical laws, thereby effectively preventing overfitting of complex models and significantly improving the physical rationality and generalization ability of the model, thereby improving the warning rationality of this monitoring and early warning system.

[0060] The specific steps of classifying the rock and soil mass into different risk levels based on the prediction results of the displacement-hydrology-geology joint prediction model and activating the corresponding level alarm are as follows:

[0061] According to the displacement curve of the rock and soil in the next 72 hours Obtain the displacement velocity curve of the rock and soil ;

[0062] According to the displacement velocity curve Obtain the maximum displacement velocity of the rock mass within the next 72 hours , based on the preset warning level classification rules and maximum displacement speed The rock and soil bodies are classified into risk levels, and the corresponding level of alarm is activated according to the classified risk level and corresponding response measures are recommended.

[0063] The preset warning level classification rules and maximum displacement speed The risk level of the rock and soil mass is classified, and the corresponding level of alarm is activated according to the risk level, and the corresponding response measures are recommended as follows:

[0064] when When the warning level is low risk, it is recommended to strengthen monitoring and pay close attention to the changing trends of monitoring data;

[0065] when When the warning level is medium risk, it is recommended to increase the monitoring frequency and prepare emergency measures;

[0066] when When the warning level is high risk, it is recommended to take emergency measures immediately and evacuate people.

[0067] It should be noted that the above steps divide the risk levels of rock and soil bodies into different levels through the early warning level classification rules, and activate the corresponding level of alarm based on the classified risk levels, which can realize graded early warning, thereby not only reminding relevant personnel to take corresponding preventive measures in advance and reasonably allocate limited resources to improve the efficiency of early warning and response, avoid waste of resources, and effectively avoid or reduce disaster losses, but also provide decision makers with a clear decision-making basis, so that they can reasonably arrange monitoring, early warning and emergency response and other tasks according to the current risk situation, optimize resource allocation, improve management efficiency, so as to ensure the scientificity and effectiveness of disaster prevention and mitigation work, and enhance the practicality of this monitoring and early warning system.

[0068] It should be understood that those skilled in the art may make improvements or modifications based on the above description, and all such improvements and modifications shall fall within the scope of protection of the appended claims. Any portion of this specification not described in detail is prior art known to those skilled in the art.

Claims

1. A monitoring and early warning system for geotechnical engineering, characterized in that: Includes: The data acquisition module is used to obtain the historical displacement data and historical hydrological data of the rock and soil body to be predicted through sensors, and obtain the historical geological data of the rock and soil body through geological exploration reports, wherein the historical geological data includes static geological data and dynamic geological data; A data processing module is used to pre-process the acquired historical displacement data, historical hydrological data and dynamic geological data to obtain pre-processed historical displacement data, historical hydrological data and dynamic geological data; The model building module is used to build a displacement-hydrology-geology coupled prediction model and incorporate the Mohr-Coulomb criterion as a regularization term into the model's loss function during model training. The prediction error of the model is calculated using the loss function, and the model is iteratively optimized through backpropagation and parameter updating. The training steps of the displacement-hydrology-geology coupled prediction model are as follows: Data preparation and preprocessing: obtaining historical displacement data, historical hydrological data, and historical geological data of the sample rock and soil mass, preprocessing all the acquired data, and dividing all the preprocessed data into training sets, validation sets, and test sets in proportion; Forward propagation and loss calculation, input the training set into the displacement-hydrology-geology coupling prediction model, and obtain the predicted displacement curve and predicted shear stress curves , according to the predicted displacement curve and actual displacement curve The difference between the two gives the displacement loss term , according to the predicted shear stress curve and ultimate shear stress curve The difference between them gives the physical law loss term based on the Mohr-Coulomb criterion , the total loss value is obtained by calculating , the calculation formula is: ,in is the regularization strength coefficient; Back propagation and parameter update, from the total loss value through the chain rule Reversely calculate the gradient of each layer's parameters and adjust the parameters using the gradient descent algorithm; Iterative optimization: divide the training set into multiple small batches, repeat the previous two steps batch by batch, and traverse the entire training set multiple times until the model converges or reaches the preset number of iterations; Verification and testing: Periodically evaluate the model's losses and indicators through the validation set to prevent overfitting. After training, evaluate the performance of the prediction model through the test set. The disaster warning module is used to input the static geological data of the rock and soil to be predicted, the pre-processed historical displacement data, the historical hydrological data and the dynamic geological data into the trained displacement-hydrological-geological joint prediction model to obtain the prediction results, which include the displacement curve of the rock and soil in the next 72 hours. and shear stress curves , based on the prediction results, the rock and soil mass is divided into different risk levels and the corresponding level of alarm is activated.

2. The monitoring and early warning system for geotechnical engineering according to claim 1, characterized in that: The historical displacement data is the displacement of the rock and soil body Time series, the historical hydrological data include instantaneous rainfall intensity time series, groundwater level height time series and pore water pressure time series, the static geological data include rock and soil type, slope, vegetation coverage, crack development index, cohesion and internal friction angle The dynamic geological data is the normal stress of the rock and soil mass Time series and shear stress Time series.

3. The monitoring and early warning system for geotechnical engineering according to claim 2, characterized in that: The displacement-hydrology-geology coupling prediction model includes three input layers, three feature extraction layers, a spatiotemporal fusion layer and two joint prediction layers, wherein the three input layers are: a time series input layer for receiving historical displacement data and dynamic geological data; a hydrological sequence input layer for receiving historical hydrological data; a geological parameter input layer for receiving static geological data; the three feature extraction layers are: an LSTM encoder layer for extracting features from the time series received by the time series input layer; a temporal convolutional network layer for extracting local features from the time series received by the hydrological sequence input layer; a fully connected embedding layer for mapping the high-dimensional geological information received by the geological parameter input layer into a low-dimensional feature vector; the spatiotemporal fusion layer is used to splice the outputs of the three feature extraction layers into a feature vector, and perform weighted fusion of the spliced ​​feature vectors through an attention mechanism to obtain a final fused feature vector; the two joint prediction layers are: a displacement prediction layer for outputting a displacement curve for the next 72 hours ; Shear stress prediction layer, used to output shear stress curves for the next 72 hours .

4. The monitoring and early warning system for geotechnical engineering according to claim 3, characterized in that: Obtain the displacement loss term The specific steps are: According to the displacement after preprocessing Time series to obtain the actual displacement curve ; The displacement loss term is obtained by calculation , the calculation formula is: ,in Represents the displacement after preprocessing The number of data in the time series, Represents the displacement after preprocessing The first The actual displacement value, Indicates the The predicted displacement value corresponding to the actual displacement value.

5. The monitoring and early warning system for geotechnical engineering according to claim 4, characterized in that: Obtain the physical law loss term The specific steps are: According to cohesion , internal friction angle and normal stress after pretreatment Time series and ultimate shear stress curve obtained by calculation , the calculation formula is: ; The physical law loss term based on the Mohr-Coulomb criterion is obtained by calculation , the calculation formula is: ,in represents the normal stress after pretreatment The number of data in the time series, Indicates the Normal stress The predicted shear stress corresponding to the data, Indicates the Normal stress The ultimate shear stress corresponding to the data is, Indicates taking and 0, whichever is greater.

6. The monitoring and early warning system for geotechnical engineering according to claim 5, characterized in that: The specific steps of classifying the rock and soil mass into different risk levels according to the prediction results and activating the corresponding level alarm are as follows: According to the displacement curve of the rock and soil in the next 72 hours Obtain the displacement velocity curve of the rock and soil ; According to the displacement velocity curve Obtain the maximum displacement velocity of the rock mass within the next 72 hours , based on the preset warning level classification rules and maximum displacement speed The rock and soil bodies are classified into risk levels, and the corresponding level of alarm is activated according to the classified risk level and corresponding response measures are recommended.

Images