Engineering geological disaster prevention and control method, device, equipment and storage medium

Abstract

The application relates to the technical field of geological disaster prevention and control, and discloses an engineering geological disaster prevention and control method, device, equipment and storage medium, wherein GNSS displacement time sequence data is acquired, a Nishihara creep model based on the creep theory of a rock-soil body is used to extract multi-scale creep stage features, an ARIMA time sequence prediction model is used to obtain a predicted displacement increment of an engineering operation area, an initial early warning threshold is dynamically modified by using an extended Kalman filter to fuse an engineering disturbance factor, and finally, a geological disaster early warning signal is generated according to the modified early warning threshold and the predicted displacement increment. Thus, the application extracts multi-scale core features directly related to the creep stage of a rock mass based on the characteristic engineering of the Nishihara creep model, accurately identifies the instability state of the rock-soil body, and then introduces an EKF mechanism fusing the engineering disturbance factor, so that the disaster early warning threshold is real-time adapted to the engineering disturbance, the false alarm and missed alarm risks are significantly reduced, and the prevention and control capability of the engineering geological disaster is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of geological disaster prevention and control technology, and in particular to a method, device, equipment and storage medium for engineering geological disaster prevention and control. Background Technology

[0002] In engineering geological disaster prevention and control, GNSS displacement data is one of the most direct and core indicators reflecting the stability of soil and rock masses, especially applicable to engineering scenarios such as mountainous highway reconstruction and expansion, and open-pit mine slopes. Activities in these scenarios generate engineering disturbances that continuously agitate the soil and rock masses, leading to gradual changes in the soil and rock masses in the engineering operation area, from slow creep to accelerated deformation and eventual instability and failure. GNSS can achieve continuous displacement monitoring with millimeter-level accuracy.

[0003] Existing early warning schemes based on GNSS data have obvious limitations: they usually only use simple time series models for trend prediction, without deeply exploring the creep stage characteristics and engineering disturbance response patterns contained in the displacement data; at the same time, the early warning threshold uses a fixed value, without considering the dynamic impact of engineering disturbances (such as the number of blasting operations per day) on the threshold, resulting in low prediction accuracy, delaying prevention and control opportunities or increasing redundant costs.

[0004] Therefore, how to improve the accuracy of engineering geological disaster prediction in the process of engineering geological disaster prevention and control is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and storage medium for preventing and controlling engineering geological disasters, aiming to solve at least one of the above-mentioned technical problems.

[0006] Optionally, to achieve the above objectives, the present invention provides a method for preventing and controlling engineering geological disasters, comprising the following steps:

[0007] The original displacement data of the engineering operation area collected by the GNSS receiver is acquired, and the original displacement data is preprocessed to obtain denoised GNSS displacement time series data.

[0008] Based on the theory of rock and soil creep, the Nishihara creep model is used to extract multi-scale creep stage characteristics.

[0009] The ARIMA time series prediction model is used, taking multi-scale creep stage characteristics and GNSS displacement time series data as input, and outputting the predicted displacement increment of the engineering operation area in the target time period.

[0010] An initial warning threshold is set by combining historical disaster data and engineering specifications, and the initial warning threshold is dynamically corrected by using extended Kalman filtering and fusing engineering disturbance factors.

[0011] Based on the revised early warning threshold and the predicted displacement increment, a geological disaster early warning signal is generated, and engineering geological disaster prevention and control is implemented based on the geological disaster early warning signal.

[0012] Optionally, the steps of acquiring raw displacement data of the engineering operation area collected by the GNSS receiver, and preprocessing the raw displacement data to obtain denoised GNSS displacement time series data specifically include:

[0013] For the RINEX 3.02 format data file output by the GNSS receiver, double-difference baseline calculation is performed using calculation software to obtain the three-direction displacement time series matrix;

[0014] The expression for the three-directional displacement time series matrix is ​​as follows: ;

[0015] In the formula, This represents a three-directional displacement time series matrix. Indicates eastward displacement with time The sequence of changes, Indicates northward displacement with time The sequence of changes, Indicates vertical displacement with time The sequence of changes, This represents the matrix transpose operation;

[0016] The three-directional displacement time series matrix is ​​smoothed by a 5-point moving average to remove high-frequency multipath noise, and then decomposed into three levels using a db4 wavelet basis. High-frequency coefficients are then processed by soft thresholding using a fixed threshold to reconstruct the denoised GNSS displacement time series.

[0017] Optionally, based on the Nishihara creep model of rock and soil creep theory, multi-scale creep stage characteristic steps are extracted, specifically including:

[0018] Based on the creep theory of soil and rock, the Nishihara creep model possesses instantaneous elastic, viscoelastic, and viscoplastic stages. Instantaneous elastic deformation features, viscoelastic creep features, viscoplastic creep features, and accelerated creep features are extracted from the denoised GNSS displacement time series, respectively. The instantaneous elastic deformation features are configured to reflect the deformation intensity of the instantaneous elastic stage, and the expression for the instantaneous elastic deformation features is as follows: ;

[0019] In the formula, The characteristic value of instantaneous elastic deformation. This represents the displacement increment within one hour after the engineering disturbance. For the moment of engineering disturbance, This is the noise-reduced displacement value one hour after the engineering disturbance. This represents the denoised displacement value at the moment of engineering disturbance;

[0020] The viscoelastic creep characteristic is configured to reflect the deformation decay rate in the viscoelastic stage, and the expression for the viscoelastic creep characteristic is as follows:

[0021] ;

[0022] ;

[0023] In the formula, This represents the characteristic value of viscoelastic creep. for Displacement value at time, Let be the fitting constant. For the initial viscoelastic displacement, The attenuation coefficient is... For time, It is a natural constant. attenuation coefficient in pass Obtained by fitting displacement data from a GNSS displacement time series;

[0024] The viscoplastic creep characteristic is configured to reflect the average deformation rate during the viscoplastic stage, and the expression for the viscoplastic creep characteristic is as follows:

[0025] ;

[0026] In the formula, Viscoplastic creep characteristic value ， for Denoising displacement value 72 hours ago For GNSS displacement time series Denoising displacement value at any time;

[0027] The accelerated creep feature is configured to reflect the displacement acceleration trend during the viscoplastic stage, and the expression for the accelerated creep feature is as follows:

[0028] ;

[0029] ;

[0030] In the formula, To accelerate creep characteristic values, for Displacement acceleration at any moment for Denoising displacement value at any time. for Denoising displacement value at any time. for Denoising displacement value at any time. The sampling interval is denoted as .

[0031] Optionally, an ARIMA time-series prediction model is used, taking multi-scale creep stage characteristics and GNSS displacement time-series data as input, to output the predicted displacement increment steps of the engineering operation area in the target time period, specifically including:

[0032] A fused input vector is constructed based on multi-scale creep stage features and denoised GNSS displacement time series; wherein, the expression of the fused input vector is specifically: ;

[0033] In the formula, This represents the fused input vector. Indicating GNSS displacement time series Denoising displacement value at any time. The characteristic value of instantaneous elastic deformation. This represents the characteristic value of viscoelastic creep. The characteristic value of viscoplastic creep is... To accelerate creep characteristic values, Transpose of a vector;

[0034] right In Perform an ADF stationarity test on the following: In , , , We performed 0-1 standardization to construct an ARIMA model that integrates features of four types of multi-scale creep stages;

[0035] The fused input vector after stationarity verification and standardization is divided into training and test sets. The least squares method is used to estimate the ARIMA model parameters, and a feature weighting term is introduced to optimize the objective function. The expression for the optimized objective function is as follows:

[0036] ;

[0037] In the formula, The optimized objective function is... for The displacement increment measured in real time over 72 hours. Based on the ARIMA model parameters p, d, q and the characteristics of four types of multi-scale creep stages , , , Calculated The displacement increment over 72 hours is given by time step, where p, d, and q represent the autoregressive order, differencing order, and moving average order of the ARIMA model, respectively. These are the characteristic weighting coefficients. for Time of the first Characteristic values ​​of multi-scale creep stages; For the first Sequence mean of characteristics of multi-scale creep stages; The total number of samples;

[0038] Iteratively train the ARIMA model until... Convergence is achieved by taking the preprocessed and fused input vectors from the past 30 days into the trained ARIMA model and outputting the displacement increment of the engineering work area over the next 72 hours. Optionally, the initial warning threshold is set by combining historical disaster data with engineering specifications, specifically including:

[0039] Historical GNSS monitoring data and four types of multi-scale creep stage characteristic data from similar projects were obtained. 200 stable samples and 30 unstable samples were selected, and a survival function incorporating the four types of multi-scale creep stage characteristics was constructed. The expression of the survival function is as follows: ;

[0040] In the formula, To integrate the survival function that incorporates the characteristics of four types of multi-scale creep stages, For displacement increment, , , , The characteristics of four types of multi-scale creep stages are as follows: , , , The characteristic influence coefficient, The scaling parameter of the Weibull distribution. The shape parameter of the Weibull distribution. It is a natural constant;

[0041] Based on engineering specifications, the stability probability of the survival function is set for different warning levels. Then, based on the stability probability of each warning level and the characteristics of four types of multi-scale creep stages, the corresponding displacement increment is solved by inverse function. According to the mapping relationship between displacement increment and warning level, four initial warning thresholds are set.

[0042] The initial warning threshold includes Blue alert Yellow alert Orange alert and A red alert is issued. Optionally, a step of dynamically correcting the initial alert threshold using an extended Kalman filter to fuse engineering perturbation factors is included, specifically:

[0043] Based on historical disturbance and instability correlation data, a mapping rule for the disturbance intensity coefficient is preset. Using real-time data of four types of multi-scale creep stage characteristics, a total disturbance coefficient integrating disturbances and characteristics is constructed. The expression for the total disturbance coefficient is as follows: ;

[0044] ;

[0045] In the formula, The total perturbation coefficient represents the fusion of perturbation and features. 0.2, 0.1, 0.3, and 0.5 represent the feature amplification coefficients based on historical data from similar projects, corresponding to instantaneous elastic features, respectively. Viscoelastic characteristics Viscoplastic characteristics Accelerated creep characteristics Sensitivity to disturbances Indicates the number of engineering disturbances. This represents the quantized value of the disturbance intensity;

[0046] Define the state vector of the extended Kalman filter, and construct the state equation of the extended Kalman filter containing nonlinear perturbation terms by combining the total perturbation coefficients. The specific expression is as follows:

[0047] ;

[0048] In the formula, , They are respectively Time threshold correction amount and rate of change of correction amount. Set to 24 hours, corresponding to a daily threshold adjustment cycle. for The total disturbance coefficient at time step is 0.12, representing the disturbance contribution coefficient, and 0.98 represents the coefficient for attenuation of the rate of change of the correction amount. The noise in the EKF process follows a mean of 0 and a covariance matrix of... The normal distribution For the diagonal matrix construction operation, 0.08 and 0.005 are the preset threshold correction amounts for equipment monitoring accuracy, respectively. Process noise standard deviation and correction rate of change The standard deviation of process noise;

[0049] Using the displacement prediction deviation at time t as the observation value of the extended Kalman filter, an extended Kalman filter observation equation containing the characteristic deviation term is constructed, and the specific expression is as follows:

[0050] ;

[0051] In the formula, for The observed value at time is equal to Predict displacement increments at all times With actual displacement increment The absolute value of the value is used to predict the displacement increment. Actual displacement increment output from the ARIMA model Data obtained through GNSS monitoring; 0.03 is the deviation coefficient. The EKF observation noise follows a pattern with a mean of 0 and a variance of . The normal distribution;

[0052] Based on the extended Kalman filter observation equation, state prediction and covariance prediction are performed in the prediction step, and Kalman gain calculation, state update, and covariance update are performed in the update step. Finally, the threshold correction amount at time t is output for threshold correction. The specific expression is as follows:

[0053] State prediction and covariance prediction in the prediction step:

[0054] ;

[0055] ;

[0056] ;

[0057] In the formula, express EKF state prediction value at time 10:00 This represents the EKF state estimate at time t-1. This represents the quantized value of the disturbance intensity. express EKF state covariance prediction at time 10:00 This represents the estimated EKF state covariance at time t-1. Represents the noise covariance matrix of the EKF process. State transition function For the state vector The partial derivative matrix, The Jacobian matrix representing the state equation;

[0058] Kalman gain calculation, state update, and covariance update in the update step:

[0059] ;

[0060] ;

[0061] ;

[0062] ;

[0063] In the formula, Indicates the EKF Kalman gain at time t. This represents the variance of the EKF observation noise. This represents the EKF state estimate at time t. express EKF observations at time 10:00 This represents the observed predicted value calculated based on the state predicted value. This represents the estimated EKF state covariance at time t. This represents the predicted value of the threshold correction. Let Jacobian matrix be the equation of observation. It is the identity matrix;

[0064] Output the threshold correction amount at time t for threshold correction:

[0065] ;

[0066] In the formula, Indicates the threshold correction amount. Indicates the initial warning threshold. This indicates the updated warning threshold. Optionally, a geological hazard warning signal is generated based on the revised warning threshold and the predicted displacement increment. Based on the geological hazard warning signal, engineering geological hazard prevention and control steps are executed, specifically including:

[0067] Displacement increments based on ARIMA model output Dynamic warning threshold after extended Kalman filter correction By comparing the data, geological disaster early warning signals of corresponding warning levels are generated;

[0068] Based on the warning level of the geological disaster warning signal, and in accordance with the pre-set prevention and control plan, engineering geological disaster prevention and control measures corresponding to the warning level are implemented.

[0069] Furthermore, to achieve the above objectives, the present invention also provides an engineering geological disaster prevention and control device, comprising:

[0070] The acquisition module is used to acquire the original displacement data of the engineering operation area collected by the GNSS receiver, and to preprocess the original displacement data to obtain the denoised GNSS displacement time series data.

[0071] The extraction module is used to extract multi-scale creep stage features from the Nishihara creep model based on the creep theory of soil and rock.

[0072] The prediction module is used to adopt the ARIMA time series prediction model, take the multi-scale creep stage characteristics and GNSS displacement time series data as input, and output the predicted displacement increment of the engineering operation area in the target time period.

[0073] The correction module is used to set the initial warning threshold by combining historical disaster data and engineering specifications, and to dynamically correct the initial warning threshold by using extended Kalman filtering and fusing engineering disturbance factors.

[0074] The execution module is used to generate a geological disaster early warning signal based on the corrected early warning threshold and the predicted displacement increment, and to perform engineering geological disaster prevention and control based on the geological disaster early warning signal.

[0075] In addition, to achieve the above objectives, the present invention also provides an engineering geological disaster prevention and control device, which includes: a memory, a processor, and an engineering geological disaster prevention and control program stored in the memory and executable on the processor. When the engineering geological disaster prevention and control program is executed by the processor, it implements the steps of the engineering geological disaster prevention and control method as described above.

[0076] In addition, to achieve the above objectives, the present invention also provides a storage medium storing an engineering geological disaster prevention and control program, which, when executed by a processor, implements the steps of the above-described engineering geological disaster prevention and control method.

[0077] The beneficial effects of this invention are as follows: It proposes a method, device, equipment, and storage medium for engineering geological disaster prevention and control. By collecting raw displacement data from the engineering operation area and processing it to obtain GNSS displacement time series data, multi-scale creep stage features are extracted based on the Nishihara creep model of rock and soil creep theory. An ARIMA time series prediction model is used to predict the predicted displacement increment of the engineering operation area during the target time period. Then, an extended Kalman filter is used to fuse engineering disturbance factors to dynamically correct the initial warning threshold. Finally, based on the corrected warning threshold and the predicted displacement increment, a geological disaster warning signal is generated to guide the implementation of engineering geological disaster prevention and control measures. Therefore, this invention, based on the characteristic engineering of the Nishihara creep model, extracts multi-scale core features directly related to the rock mass creep stage, accurately identifies the instability state of the rock and soil mass, and introduces an EKF mechanism that fuses engineering disturbance factors, enabling the disaster warning threshold to adapt to engineering disturbances in real time, significantly reducing the risk of false alarms and missed alarms, and effectively improving the prevention and control capabilities of engineering geological disasters. Attached Figure Description

[0078] Figure 1This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention;

[0079] Figure 2 This is a flowchart illustrating an embodiment of the engineering geological disaster prevention and control method of the present invention;

[0080] Figure 3 This is a structural block diagram of an engineering geological disaster prevention and control device according to an embodiment of the present invention. Detailed Implementation

[0081] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0082] like Figure 1 As shown, Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention.

[0083] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU; a communication bus 1002; a user interface 1003; a network interface 1004; and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0084] Those skilled in the art will understand that Figure 1 The structure of the device shown does not constitute a limitation on the device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0085] like Figure 1 As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an engineering geological disaster prevention and control program.

[0086] exist Figure 1In the terminal shown, network interface 1004 is mainly used to connect to the backend server and communicate data with it; user interface 1003 is mainly used to connect to the client (user terminal) and communicate data with it; while processor 1001 can be used to call the engineering geological disaster prevention and control program stored in memory 1005 and perform the following operations:

[0087] The original displacement data of the engineering operation area collected by the GNSS receiver is acquired, and the original displacement data is preprocessed to obtain denoised GNSS displacement time series data.

[0088] Based on the theory of rock and soil creep, the Nishihara creep model is used to extract multi-scale creep stage characteristics.

[0089] The ARIMA time series prediction model is used, taking multi-scale creep stage characteristics and GNSS displacement time series data as input, and outputting the predicted displacement increment of the engineering operation area in the target time period.

[0090] An initial warning threshold is set by combining historical disaster data and engineering specifications, and the initial warning threshold is dynamically corrected by using extended Kalman filtering and fusing engineering disturbance factors.

[0091] Based on the revised early warning threshold and the predicted displacement increment, a geological disaster early warning signal is generated, and engineering geological disaster prevention and control is implemented based on the geological disaster early warning signal.

[0092] The specific embodiments of the present invention applied to the device are basically the same as the embodiments of the engineering geological disaster prevention and control methods described below, and will not be repeated here.

[0093] This invention provides a method for preventing and controlling engineering geological disasters, referring to... Figure 2 , Figure 2 This is a flowchart illustrating an embodiment of the engineering geological disaster prevention and control method of the present invention.

[0094] In this embodiment, a method for preventing and controlling engineering geological disasters includes the following steps:

[0095] S100: Acquire the original displacement data of the engineering operation area collected by the GNSS receiver, preprocess the original displacement data, and obtain the denoised GNSS displacement time series data.

[0096] S200: The Nishihara creep model based on the theory of rock and soil creep, extracting multi-scale creep stage characteristics;

[0097] S300: Using the ARIMA time series prediction model, the multi-scale creep stage characteristics and GNSS displacement time series data are used as inputs to output the predicted displacement increment of the engineering operation area in the target time period.

[0098] S400: The initial warning threshold is set by combining historical disaster data and engineering specifications, and the initial warning threshold is dynamically corrected by using extended Kalman filter to fuse engineering disturbance factors;

[0099] S500: Generate a geological disaster early warning signal based on the corrected early warning threshold and the predicted displacement increment, and perform engineering geological disaster prevention and control based on the geological disaster early warning signal.

[0100] It should be noted that existing GNSS data-based early warning schemes have significant limitations: they typically only use simple time-series models for trend prediction, failing to delve into the creep stage characteristics and engineering disturbance response patterns inherent in the displacement data; furthermore, they use fixed early warning thresholds, neglecting the dynamic impact of engineering disturbances (such as the number of blasting operations per day) on the thresholds, resulting in low prediction accuracy, delayed prevention and control, or increased redundant costs. To address these issues, this embodiment collects raw displacement data from the engineering operation area, processes it to obtain GNSS displacement time-series data, extracts multi-scale creep stage characteristics based on the Nishihara creep model of soil and rock creep theory, uses the ARIMA time-series prediction model to predict the predicted displacement increment of the engineering operation area during the target period, then uses extended Kalman filtering to fuse engineering disturbance factors to dynamically correct the initial early warning threshold, and finally generates a geological disaster early warning signal based on the corrected early warning threshold and the predicted displacement increment to guide the implementation of engineering geological disaster prevention and control measures. Therefore, based on the characteristic engineering of the Nishihara creep model, this invention extracts multi-scale core features directly related to the creep stage of the rock mass, accurately identifies the instability state of the rock and soil mass, and introduces the EKF mechanism that integrates engineering disturbance factors, so that the disaster early warning threshold can be adapted to engineering disturbance in real time, significantly reducing the risk of false alarms and missed alarms, and effectively improving the prevention and control capabilities of engineering geological disasters.

[0101] In a preferred embodiment, the steps of acquiring raw displacement data of the engineering operation area collected by a GNSS receiver, and preprocessing the raw displacement data to obtain denoised GNSS displacement time series data specifically include:

[0102] S110: For the RINEX 3.02 format data file output by the GNSS receiver, double-difference baseline calculation is performed using calculation software to obtain the three-direction displacement time series matrix;

[0103] The expression for the three-directional displacement time series matrix is ​​as follows:

[0104] ;

[0105] In the formula, This represents a three-directional displacement time series matrix. Indicates eastward displacement with time The sequence of changes, Indicates northward displacement with time The sequence of changes, Indicates vertical displacement with time The sequence of changes, This represents the matrix transpose operation;

[0106] S120: For the three-direction displacement time series matrix, high-frequency multipath noise is smoothed by 5-point moving average, and then 3-level decomposition is performed using db4 wavelet basis. High-frequency coefficient soft thresholding is performed using fixed threshold to reconstruct the denoised GNSS displacement time series.

[0107] In this embodiment, for the RINEX 3.02 format data output by the GNSS receiver, a three-direction displacement sequence is obtained by double-difference baseline calculation, and then denoised by a combination of moving average and wavelet thresholding to obtain a high-quality GNSS displacement time series.

[0108] Specifically, it includes the following execution process: First, format decomposition and sequence construction: using GAMIT / GLOBK10.7 software to perform double-difference baseline decomposition on RINEX 3.02 format data to generate eastward-oriented... North Vertical Three-direction displacement time series matrix The second step is noise suppression. First, high-frequency multipath noise is smoothed by a 5-point moving average. Then, a 3-level decomposition is performed using the db4 wavelet basis. Soft thresholding is applied to the high-frequency detail coefficients with a fixed threshold. After reconstruction, the denoised displacement time series is obtained.

[0109] In a preferred embodiment, the Nishihara creep model based on the theory of soil and rock creep extracts multi-scale creep stage characteristic steps, specifically including:

[0110] S210: The instantaneous elastic stage, viscoelastic stage and viscoplastic stage of the Nishihara creep model based on the creep theory of soil and rock are extracted from the denoised GNSS displacement time series, respectively.

[0111] The instantaneous elastic deformation feature is configured to reflect the deformation intensity during the instantaneous elastic phase, and the expression for the instantaneous elastic deformation feature is specifically as follows:

[0112] ;

[0113] In the formula, The characteristic value of instantaneous elastic deformation. This represents the displacement increment within one hour after the engineering disturbance. For the moment of engineering disturbance, This is the noise-reduced displacement value one hour after the engineering disturbance. This represents the denoised displacement value at the moment of engineering disturbance;

[0114] The viscoelastic creep characteristic is configured to reflect the deformation decay rate in the viscoelastic stage, and the expression for the viscoelastic creep characteristic is as follows:

[0115] ;

[0116] ;

[0117] In the formula, This represents the characteristic value of viscoelastic creep. for Displacement value at time, Let be the fitting constant. For the initial viscoelastic displacement, The attenuation coefficient is... For time, It is a natural constant. attenuation coefficient in pass Obtained by fitting displacement data from a GNSS displacement time series;

[0118] The viscoplastic creep characteristic is configured to reflect the average deformation rate during the viscoplastic stage, and the expression for the viscoplastic creep characteristic is as follows: ;

[0119] In the formula, Viscoplastic creep characteristic value ， for Denoising displacement value 72 hours ago For GNSS displacement time series Denoising displacement value at any time;

[0120] The accelerated creep feature is configured to reflect the displacement acceleration trend during the viscoplastic stage, and the expression for the accelerated creep feature is as follows: ;

[0121] ;

[0122] In the formula, To accelerate creep characteristic values, for Displacement acceleration at any moment for Denoising displacement value at any time. for Denoising displacement value at any time. for Denoising displacement value at any time. The sampling interval is denoted as .

[0123] In this embodiment, based on the three-stage characteristics of the Nishihara creep model—instantaneous elasticity, viscoelasticity, and viscoplasticity—and grounded in the fundamental nature of soil and rock creep mechanics, features are extracted in stages: First, the stage patterns of the Nishihara model are clarified: instantaneous elastic stage (displacement increment concentrated within 1 hour after disturbance), viscoelastic stage (deformation decay within 1-24 hours after disturbance), and viscoplastic stage (accelerated deformation more than 72 hours after disturbance). Second, four types of features are extracted specifically from the denoised GNSS displacement sequence: instantaneous elasticity, viscoelasticity, viscoplasticity, and accelerated creep. Thus, by extracting these four types of features directly related to the creep stages of soil and rock, the precursors to soil and rock instability can be accurately identified, improving the accuracy of engineering geological hazard prediction and reducing false alarm and false negative rates.

[0124] In a preferred embodiment, the ARIMA time-series prediction model is used, taking multi-scale creep stage characteristics and GNSS displacement time-series data as input, and outputting the predicted displacement increment of the engineering operation area in the target time period, specifically including:

[0125] S310: Based on the multi-scale creep stage characteristics and the denoised GNSS displacement time series, a fused input vector is constructed; wherein, the expression of the fused input vector is specifically: ;

[0126] In the formula, This represents the fused input vector. Indicating GNSS displacement time series Denoising displacement value at any time. The characteristic value of instantaneous elastic deformation. This represents the characteristic value of viscoelastic creep. The characteristic value of viscoplastic creep is... To accelerate creep characteristic values, Transpose of a vector;

[0127] S320: Yes In Perform an ADF stationarity test on the following: In , , , Perform 0-1 standardization; construct an ARIMA model that integrates features of four types of multi-scale creep stages;

[0128] S330: The fused input vector after stationarity verification and standardization is divided into training and test sets. The least squares method is used to estimate the ARIMA model parameters, and a feature weighting term is introduced to optimize the objective function. The expression of the optimized objective function is as follows: ;

[0129] In the formula, The optimized objective function is... for The displacement increment measured in real time over 72 hours. Based on the ARIMA model parameters p, d, q and the characteristics of four types of multi-scale creep stages , , , Calculated The displacement increment over 72 hours is given by time step, where p, d, and q represent the autoregressive order, differencing order, and moving average order of the ARIMA model, respectively. These are the characteristic weighting coefficients. for Time of the first Characteristic values ​​of multi-scale creep stages; For the first Sequence mean of characteristics of multi-scale creep stages; The total number of samples;

[0130] S340: Iteratively train the ARIMA model until... Convergence is achieved by taking the preprocessed and fused input vectors from the past 30 days into the trained ARIMA model and outputting the displacement increment of the engineering work area over the next 72 hours. .

[0131] In this embodiment, an input vector is constructed by fusing denoised GNSS displacement sequences and multi-scale creep features. An ARIMA model (considering stationarity processing, standardization, and objective function optimization) is then used to predict the displacement increment of the engineering work area for a future target period. Therefore, by fusing creep features, the model can predict the displacement increment of the soil and rock mass based on its stage state (e.g., ...). The dynamic adjustment of prediction logic (indicating accelerated viscoplasticity) significantly improves prediction accuracy, overcoming the shortcomings of traditional ARIMA models that rely solely on displacement time and ignore the physical state of soil and rock. This avoids misjudging construction vibrations as instability or missing slow acceleration instability due to neglecting physical meaning, thus improving prediction accuracy.

[0132] In a preferred embodiment, the step of setting an initial early warning threshold by combining historical disaster data with engineering specifications specifically includes:

[0133] S410: Obtain historical GNSS monitoring data and four types of multi-scale creep stage characteristic data from similar projects, screen 200 stable samples and 30 unstable samples, and construct a survival function incorporating the four types of multi-scale creep stage characteristics; wherein, the expression of the survival function is as follows: ;

[0134] In the formula, To integrate the survival function that incorporates the characteristics of four types of multi-scale creep stages, For displacement increment, , , , The characteristics of four types of multi-scale creep stages are as follows: , , , The characteristic influence coefficient, The scaling parameter of the Weibull distribution. The shape parameter of the Weibull distribution. It is a natural constant;

[0135] S420: Combining engineering specifications, set the stability probability of the survival function for different warning levels, and then, based on the stability probability of each warning level and the characteristics of four types of multi-scale creep stages, solve the corresponding displacement increment through the inverse function. According to the mapping relationship between the displacement increment and the warning level, set the initial warning threshold for four levels.

[0136] The initial warning threshold includes Blue alert Yellow alert Orange alert and A red alert has been issued.

[0137] Based on this, the initial warning threshold is dynamically corrected by fusing the engineering perturbation factor using an extended Kalman filter, specifically including:

[0138] S430: Based on historical engineering disturbance and instability correlation data, a mapping rule for the disturbance intensity coefficient is preset. Using real-time data of four types of multi-scale creep stage characteristics, a total disturbance coefficient integrating disturbances and characteristics is constructed. The expression for the total disturbance coefficient is as follows: ;

[0139] ;

[0140] In the formula, The total perturbation coefficient represents the fusion of perturbation and features. 0.2, 0.1, 0.3, and 0.5 represent the feature amplification coefficients based on historical data from similar projects, corresponding to instantaneous elastic features, respectively. Viscoelastic characteristics Viscoplastic characteristics Accelerated creep characteristics Sensitivity to disturbances Indicates the number of engineering disturbances. This represents the quantized value of the disturbance intensity;

[0141] S440: Define the state vector of the extended Kalman filter, and construct the state equation of the extended Kalman filter containing nonlinear perturbation terms by combining the total perturbation coefficients. The specific expression is as follows; ;

[0142] In the formula, , They are respectively Time threshold correction amount and rate of change of correction amount. Set to 24 hours, corresponding to a daily threshold adjustment cycle. for The total disturbance coefficient at time step is 0.12, representing the disturbance contribution coefficient, and 0.98 represents the coefficient for attenuation of the rate of change of the correction amount. The noise in the EKF process follows a mean of 0 and a covariance matrix of... The normal distribution For the diagonal matrix construction operation, 0.08 and 0.005 are the preset threshold correction amounts for equipment monitoring accuracy, respectively. Process noise standard deviation and correction rate of change The standard deviation of process noise;

[0143] S450: will The displacement prediction error at time t is used as the observation value of the extended Kalman filter. The extended Kalman filter observation equation containing the characteristic deviation term is constructed, and the specific expression is as follows: ;

[0144] In the formula, for The observed value at time is equal to Predict displacement increments at all times With actual displacement increment The absolute value of the value is used to predict the displacement increment. Actual displacement increment output from the ARIMA model Data obtained through GNSS monitoring; 0.03 is the deviation coefficient. The EKF observation noise follows a pattern with a mean of 0 and a variance of . The normal distribution;

[0145] S460: Based on the extended Kalman filter observation equation, perform state prediction and covariance prediction in the prediction step, and perform Kalman gain calculation, state update, and covariance update in the update step. Finally, output the threshold correction amount at time t for threshold correction. The specific expression is: State prediction and covariance prediction in the prediction step: ;

[0146] ;

[0147] ;

[0148] In the formula, express EKF state prediction value at time 10:00 This represents the EKF state estimate at time t-1. This represents the quantized value of the disturbance intensity. This represents the predicted EKF state covariance value at time t. This represents the estimated EKF state covariance at time t-1. Represents the noise covariance matrix of the EKF process. State transition function For the state vector The partial derivative matrix, The Jacobian matrix representing the state equation;

[0149] Kalman gain calculation, state update, and covariance update in the update step: ;

[0150] ;

[0151] ;

[0152] ;

[0153] In the formula, Indicates the EKF Kalman gain at time t. This represents the variance of the EKF observation noise. This represents the EKF state estimate at time t. Represents the EKF observation value at time t. This represents the observed predicted value calculated based on the state predicted value. This represents the estimated EKF state covariance at time t. This represents the predicted value of the threshold correction. Let Jacobian matrix be the equation of observation. It is the identity matrix;

[0154] Output the threshold correction amount at time t for threshold correction: ;

[0155] In the formula, Indicates the threshold correction amount. Indicates the initial warning threshold. This indicates the updated warning threshold.

[0156] In this embodiment, a total disturbance coefficient is constructed based on engineering disturbance factors (such as the number of blastings per day) and multi-scale creep characteristics. A nonlinear state observation equation is constructed through extended Kalman filtering to achieve dynamic correction of the initial warning threshold.

[0157] Specifically, it includes the following execution process: First, the disturbance factor is quantified and the total disturbance coefficient is calculated, including the number of blasts per day. Quantified as disturbance intensity coefficient The total perturbation coefficient is constructed by combining creep characteristics. The second step involves constructing and defining the EKF model, including the state vector, state equation, and observation equation. The third step is the iterative update of the EKF model: the prediction step calculates the predicted values ​​of the state and covariance, and the update step calculates the updated values ​​of the Kalman gain, state, and covariance. Finally, a correction value is output to adjust the dynamic threshold. This allows the early warning threshold to adapt in real-time to changes in engineering disturbances and the evolution of soil and rock conditions. By handling nonlinear disturbances through EKF, the threshold correction value can quickly respond to changes such as increased blasting frequency, preventing threshold failure during sudden disturbances and significantly reducing false alarms and false negatives.

[0158] In a preferred embodiment, a geological disaster early warning signal is generated based on the corrected early warning threshold and the predicted displacement increment. Based on the geological disaster early warning signal, engineering geological disaster prevention and control steps are executed, specifically including:

[0159] S510: Displacement increment based on ARIMA model output Dynamic warning threshold after extended Kalman filter correction By comparing the data, geological disaster early warning signals of corresponding warning levels are generated;

[0160] S520: Based on the warning level of the geological disaster warning signal, and in accordance with the pre-set prevention and control plan, implement engineering geological disaster prevention and control measures corresponding to the warning level.

[0161] In this embodiment, by comparing the predicted displacement increment output by the ARIMA model with the dynamic threshold corrected by EKF, a corresponding level of early warning signal is generated, and engineering geological disaster prevention and control measures are implemented according to the preset prevention and control plan. Therefore, a high-precision engineering geological disaster prevention and control method based on GNSS displacement data is proposed. This method extracts stage features of the soil and rock mass based on the Nishihara creep model to give the prediction physical meaning, utilizes ARIMA to fuse features to improve the accuracy of displacement increment prediction, and then introduces EKF to combine disturbances and features to achieve dynamic threshold adaptation. Finally, through multi-dimensional decision-making, early warnings are generated and linked with prevention and control, enabling the disaster early warning threshold to adapt to engineering disturbances in real time, significantly reducing the risk of false alarms and missed alarms, and effectively improving the prevention and control capabilities of engineering geological disasters.

[0162] Reference Figure 3 , Figure 3 This is a structural block diagram of an embodiment of the engineering geological disaster prevention and control device of the present invention.

[0163] like Figure 3 As shown, the engineering geological disaster prevention and control device proposed in this embodiment of the invention includes:

[0164] The acquisition module 10 is used to acquire the original displacement data of the engineering operation area collected by the GNSS receiver, and to preprocess the original displacement data to obtain the denoised GNSS displacement time series data.

[0165] Extraction module 20 is used to extract multi-scale creep stage features from the Nishihara creep model based on the creep theory of soil and rock.

[0166] Prediction module 30 is used to adopt the ARIMA time series prediction model, take multi-scale creep stage characteristics and GNSS displacement time series data as input, and output the predicted displacement increment of the engineering operation area in the target time period.

[0167] The correction module 40 is used to set the initial warning threshold by combining historical disaster data and engineering specifications, and to dynamically correct the initial warning threshold by using extended Kalman filtering and fusing engineering disturbance factors.

[0168] The execution module 50 is used to generate a geological disaster early warning signal based on the corrected early warning threshold and the predicted displacement increment, and to perform engineering geological disaster prevention and control based on the geological disaster early warning signal.

[0169] Other embodiments or specific implementations of the engineering geological disaster prevention and control device of the present invention can be referred to the above-described method embodiments, and will not be repeated here.

[0170] Furthermore, the present invention also proposes an engineering geological disaster prevention and control device, which includes: a memory, a processor, and an engineering geological disaster prevention and control program stored in the memory and executable on the processor. When the engineering geological disaster prevention and control program is executed by the processor, it implements the steps of the engineering geological disaster prevention and control method described above.

[0171] The specific implementation method of the engineering geological disaster prevention and control equipment in this application is basically the same as the embodiments of the above-mentioned engineering geological disaster prevention and control methods, and will not be repeated here.

[0172] Furthermore, this invention also proposes a readable storage medium, which includes a computer-readable storage medium storing an engineering geological disaster prevention and control program thereon. The readable storage medium may be... Figure 1 The memory 1005 in the terminal can also be at least one of ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk, etc. The readable storage medium includes several instructions to cause an engineering geological disaster prevention and control device with a processor to execute the engineering geological disaster prevention and control method described in the various embodiments of the present invention.

[0173] The specific implementation methods in the readable storage medium of this application are basically the same as those in the embodiments of the above-described engineering geological disaster prevention and control methods, and will not be described again here.

[0174] It is understood that in the description of this specification, references to terms such as "one embodiment," "another embodiment," "other embodiments," or "first embodiment to Nth embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0175] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0176] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0177] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0178] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. An engineering geological disaster prevention and control method, characterized in that, The method comprises the following steps: Obtaining original displacement data of an engineering operation area collected by a GNSS receiver, preprocessing the original displacement data, and obtaining denoised GNSS displacement time series data; Based on the Nishihara creep model of the creep theory of rock-soil mass, multi-scale creep stage features are extracted; Specifically comprising: Based on the instantaneous elastic stage, viscoelastic stage and viscoplastic stage of the Nishihara creep model of the creep theory of rock-soil mass, instantaneous elastic deformation features, viscoelastic creep features, viscoplastic creep features and accelerated creep features are extracted from the denoised GNSS displacement time series respectively. The instantaneous elastic deformation features are configured to reflect the deformation intensity in the instantaneous elastic stage, and the expression of the instantaneous elastic deformation features is specifically as follows: ； In the formula, is the instantaneous elastic deformation characteristic value, is the displacement increment within 1 hour after the engineering disturbance, is the engineering disturbance time, is the denoised displacement value 1 hour after the engineering disturbance, is the denoised displacement value at the engineering disturbance time; The viscoelastic creep features are configured to reflect the deformation decay rate in the viscoelastic stage, and the expression of the viscoelastic creep features is specifically as follows: ； ； In the formula, This is a characteristic value of viscoelastic creep. for Displacement value at time, Let be the fitting constant. For the initial viscoelastic displacement, The attenuation coefficient is... For time, It is a natural constant. attenuation coefficient in pass Obtained by fitting displacement data from a GNSS displacement time series; The viscoplastic creep features are configured to reflect the average deformation rate in the viscoplastic stage, and the expression of the viscoplastic creep features is specifically as follows: ； In the formula, is a viscous plastic creep characteristic value ， is the denoised displacement value at time 72 hours ago, is the GNSS displacement time series the denoised displacement value at time The accelerated creep features are configured to reflect the displacement acceleration trend in the viscoplastic stage, and the expression of the accelerated creep features is specifically as follows: ； ； In the formula, for accelerating the creep characteristic value, for time displacement acceleration, for time denoised displacement value, for time denoised displacement value, for time denoised displacement value, is a sampling interval; An ARIMA time series prediction model is adopted, the multi-scale creep stage features and the GNSS displacement time series data are taken as inputs, and the predicted displacement increment of the engineering operation area in a target period is output. An initial warning threshold is set in combination with historical disaster data and engineering specifications, and the initial warning threshold is dynamically modified by using an extended Kalman filter to fuse engineering disturbance factors. According to the modified warning threshold and the predicted displacement increment, a geological disaster warning signal is generated, and engineering geological disaster prevention and control is performed based on the geological disaster warning signal.

2. The method of claim 1, wherein the method comprises, The step of obtaining original displacement data of an engineering operation area collected by a GNSS receiver, preprocessing the original displacement data, and obtaining denoised GNSS displacement time series data comprises: For the RINEX3.02 format data file output by the GNSS receiver, double-difference baseline solution is performed through a solution software to obtain a three-direction displacement time series matrix. The expression of the three-direction displacement time series matrix is specifically as follows: ； In the formula, This represents a three-directional displacement time series matrix. Indicates eastward displacement with time The sequence of changes, Indicates northward displacement with time The sequence of changes, Indicates vertical displacement with time The sequence of changes, This represents the matrix transpose operation; The three-direction displacement time series matrix is smoothed by 5-point sliding average to smooth high-frequency multipath noise, and then is decomposed by db4 wavelet basis for 3 layers, and high-frequency coefficient soft threshold processing is performed by using a fixed threshold to obtain denoised GNSS displacement time series through reconstruction.

3. The method of claim 1, wherein the method comprises, The step of adopting an ARIMA time series prediction model, taking the multi-scale creep stage features and the GNSS displacement time series data as inputs, and outputting the predicted displacement increment of the engineering operation area in a target period comprises: Based on the multi-scale creep stage features and the denoised GNSS displacement time series, a fusion input vector is constructed, and the expression of the fusion input vector is specifically as follows: ； wherein, denotes the fused input vector, denotes the GNSS displacement time series, denotes the denoised displacement value at time instant, is the instantaneous elastic deformation eigenvalue, is the visco-elastic creep eigenvalue, is the visco-plastic creep eigenvalue, is the accelerated creep eigenvalue, is the vector transpose; right In Perform an ADF stationarity test on the following: In , , , We performed 0-1 standardization to construct an ARIMA model that integrates features of four types of multi-scale creep stages; The stationary check and the standardized fusion input vector are divided into a training set and a test set, the least square method is used to estimate the ARIMA model parameters, and a feature weighting term is introduced to optimize the objective function; wherein the expression of the optimized objective function is: In the formula, The optimized objective function is... for The displacement increment measured in real time over 72 hours. Based on the ARIMA model parameters p, d, q and the characteristics of four types of multi-scale creep stages , , , Calculated The displacement increment over 72 hours is given by time step, where p, d, and q represent the autoregressive order, differencing order, and moving average order of the ARIMA model, respectively. These are the characteristic weighting coefficients. for Time of the first Characteristic values ​​of multi-scale creep stages; For the first Sequence mean of characteristics of multi-scale creep stages, The total number of samples; Iteratively train the ARIMA model until convergence, by the trained ARIMA model, input the pre-processed fused input vector of the last 30 days, output the displacement increment of the engineering work area in the next 72 hours .

4. The method of claim 1, wherein the method comprises, In combination with historical disaster data and engineering specifications, the initial warning threshold is set in steps, specifically including: Obtain GNSS monitoring historical data and 4 types of multi-scale creep stage characteristic data of similar projects, select 200 stable samples and 30 unstable samples, and construct a survival function incorporating 4 types of multi-scale creep stage characteristics; wherein the expression of the survival function is: ； wherein, is the survival function fusing the features of the four classes of multi-scale creep stages, is the displacement increment, , , , is the feature of the four classes of multi-scale creep stages, , , , is the feature influence coefficient, is the Weibull distribution scale parameter, is the Weibull distribution shape parameter, is the natural constant; In combination with engineering specifications, set the stability probability of the survival function for different warning levels, and based on the stability probability of each warning level and the 4 types of multi-scale creep stage characteristics, solve the corresponding displacement increment by inverse function, and set the 4-level initial warning threshold according to the mapping relationship between the displacement increment and the warning level; wherein the initial warning threshold comprises a blue warning, a yellow warning, an orange warning, and a red warning.

5. The method of claim 4, wherein the method further comprises: The initial warning threshold is dynamically modified by using an extended Kalman filter to fuse engineering disturbance factors in steps, specifically including: According to the mapping rule of disturbance intensity coefficient based on the correlation data of engineering historical disturbance and instability, the total disturbance coefficient is constructed by fusing disturbance and characteristics according to real-time data of 4 types of multi-scale creep stage characteristics; wherein the expression of the total disturbance coefficient is: ； ； In the formula, represents the total disturbance coefficient of fusion disturbance and characteristics, 0.2, 0.1, 0.3, 0.5 represent the characteristic amplification coefficient fitted based on the historical data of the same kind of engineering, respectively corresponding to the instantaneous elastic characteristic , the viscoelastic characteristic , the viscoplastic characteristic , the accelerated creep characteristic sensitivity to disturbance, represents the number of engineering disturbances, represents the disturbance intensity quantitative value; The state vector of the extended Kalman filter is defined, and the state equation of the extended Kalman filter containing the nonlinear disturbance term is constructed in combination with the total disturbance coefficient, and the expression is specifically as follows: ; In the formula, , respectively, The time threshold correction amount and the correction amount change rate, Set to 24 hours, corresponding to the threshold correction period of once a day, The total disturbance coefficient at time t is 0.12, and the disturbance contribution coefficient is 0.

98. EKF process noise, obeying normal distribution with mean value 0 and covariance matrix , Diagonal matrix construction operation, 0.08 and 0.005 are respectively the process noise standard deviation of the threshold correction amount preset by the equipment monitoring accuracy And the process noise standard deviation of the correction amount change rate ; the displacement prediction deviation at time t is taken as the observation value of the extended Kalman filter, and the extended Kalman filter observation equation containing the characteristic deviation term is constructed, and the expression is specifically:​ ； In the formula, is the observation value at time t, equal to the predicted displacement increment at time t and the absolute value of the actual displacement increment the predicted displacement increment output by the ARIMA model, the actual displacement increment obtained through GNSS monitoring, and 0.03 is a bias coefficient, is the EKF observation noise, subject to a normal distribution with a mean of 0 and a variance of According to the extended Kalman filter observation equation, the state prediction and covariance prediction of the prediction step are performed, the Kalman gain calculation, state update and covariance update of the update step are performed, and finally the threshold correction amount at time t is output to perform threshold correction. The expression is specifically: State prediction and covariance prediction of the prediction step: ； ； ； wherein denotes the EKF state prediction at time t, denotes the EKF state estimate at time t-1, denotes the disturbance intensity quantification, denotes the EKF state covariance prediction at time t, denotes the EKF state covariance estimate at time t-1, denotes the EKF process noise covariance matrix, denotes the state transition function denotes the partial derivative matrix of the state vector with respect to the state vector denotes the Jacobian matrix of the state equation; Kalman gain computation, state update and covariance update of the update step: ； ； ； ； wherein denotes the EKF Kalman gain at time t, denotes the EKF observation noise variance, denotes the EKF state estimate at time t, denotes the EKF observation at time t, denotes the observation prediction based on the state prediction, denotes the EKF state covariance estimate at time t, denotes the threshold correction amount prediction, is the Jacobian matrix of the observation equation, is the identity matrix; outputs the threshold correction amount at time t for threshold correction: ； In the formula, represents a threshold correction amount, represents an initial early warning threshold value, represents an updated early warning threshold value.

6. The method of claim 5, wherein the method further comprises: According to the modified warning threshold and the predicted displacement increment, a geological disaster warning signal is generated, and based on the geological disaster warning signal, an engineering geological disaster prevention and control step is performed, specifically including: Displacement increments based on arima model output Dynamic early warning threshold and extended kalman filter correction Conduct comparison, generate geological disaster early warning signal corresponding to early warning level; According to the warning level of the geological disaster warning signal, the corresponding engineering geological disaster prevention and control measures are performed according to the pre-set prevention and control plan.

7. An engineering geological disaster prevention and control device, characterized in that, Including: An acquisition module is configured to acquire original displacement data of an engineering operation area collected by a GNSS receiver, and to obtain denoised GNSS displacement time series data by preprocessing the original displacement data; An extraction module is configured to extract multi-scale creep stage characteristics based on the Nishihara creep model of rock mass creep theory; specifically including: Based on the instantaneous elastic stage, viscoelastic stage and viscoplastic stage of the Nishihara creep model of rock mass creep theory, instantaneous elastic deformation characteristics, viscoelastic creep characteristics, viscoplastic creep characteristics and accelerated creep characteristics are extracted from the denoised GNSS displacement time series; Wherein, the instantaneous elastic deformation characteristics are configured to reflect the deformation intensity in the instantaneous elastic stage, and the expression of the instantaneous elastic deformation characteristics is: ； In the formula, is the instantaneous elastic deformation characteristic value, is the displacement increment within 1 hour after the engineering disturbance, is the time of the engineering disturbance, is the denoised displacement value 1 hour after the engineering disturbance, is the denoised displacement value at the time of the engineering disturbance; Wherein, the viscoelastic creep characteristics are configured to reflect the deformation decay rate in the viscoelastic stage, and the expression of the viscoelastic creep characteristics is: ； ； In the formula, This is a characteristic value of viscoelastic creep. for Displacement value at time, Let be the fitting constant. For the initial viscoelastic displacement, The attenuation coefficient is... For time, It is a natural constant. attenuation coefficient in pass Obtained by fitting displacement data from a GNSS displacement time series; Wherein, the viscoplastic creep characteristics are configured to reflect the average deformation rate in the viscoplastic stage, and the expression of the viscoplastic creep characteristics is: ； In the formula, is the viscoplastic creep characteristic value ， is is the denoised displacement value at time 72 hours ago, is the GNSS displacement time series is the denoised displacement value at time Wherein, the accelerated creep characteristics are configured to reflect the displacement acceleration trend in the viscoplastic stage, and the expression of the accelerated creep characteristics is: ； ； In the formula, for accelerating the creep characteristic value, for time displacement acceleration, for time denoised displacement value, for time denoised displacement value, for time denoised displacement value, is a sampling interval; The prediction module is configured to use an ARIMA time series prediction model, take the multi-scale creep stage features and GNSS displacement time series data as input, and output a predicted displacement increment of the engineering operation area in a target period; The correction module is configured to set an initial early warning threshold in combination with historical disaster data and engineering specifications, and dynamically correct the initial early warning threshold by using an extended Kalman filter to fuse engineering disturbance factors; The execution module is configured to generate a geological disaster early warning signal according to the corrected early warning threshold and the predicted displacement increment, and perform engineering geological disaster prevention and control based on the geological disaster early warning signal.

8. An engineering geological disaster prevention and control equipment, characterized in that, The engineering geological disaster prevention and control device comprises a memory, a processor, and an engineering geological disaster prevention and control program stored on the memory and executable on the processor. When the engineering geological disaster prevention and control program is executed by the processor, the steps of the engineering geological disaster prevention and control method according to any one of claims 1 to 6 are implemented.

9. A storage medium, characterized by The storage medium stores an engineering geological disaster prevention and control program. When the engineering geological disaster prevention and control program is executed by the processor, the steps of the engineering geological disaster prevention and control method according to any one of claims 1 to 6 are implemented.

Images