Reservoir bank slope deformation monitoring and early warning processing method based on InSAR (Interferometric Synthetic Aperture Radar) data

Abstract

The invention discloses a reservoir bank slope deformation monitoring and early warning processing method based on InSAR data, and relates to the technical field of reservoir bank slope deformation monitoring, and the method specifically comprises the following steps: obtaining slope deformation characteristic information influenced by sudden drop of a water level in real time under the condition that the water level of a reservoir is determined to drop sharply in a short time, and carrying out the analysis after obtaining, respectively generating a water level sudden drop deformation response coefficient and an instantaneous instability trend index; constructing an instantaneous instability risk assessment model for the generated water level sudden drop deformation response coefficient and the instantaneous instability trend index, and generating a slope instantaneous instability assessment coefficient; and performing evaluation analysis based on the slope instantaneous instability evaluation coefficient, and evaluating the slope instantaneous instability risk under the condition that the water level of the reservoir drastically drops within a short time. According to the method, the problem that the instantaneous instability risk of the side slope cannot be evaluated under the condition that the water level of the reservoir drops sharply within a short time in the existing InSAR monitoring technology is solved, and dynamic monitoring and early warning with high timeliness and accurate quantification are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of deformation monitoring of reservoir bank slopes, and particularly relates to a method for processing deformation monitoring and early warning of reservoir bank slopes based on InSAR data. Background Art

[0002] Reservoir bank slopes refer to the slope-like landforms formed along the coasts of large reservoirs or rivers due to the differences in terrain undulation and geological conditions. Under the long-term changes in hydrological conditions and the interference of human activities, these slopes are prone to landslides, collapses or other geological disasters, posing a serious threat to the lives of downstream residents, infrastructure and ecological environment. Since the stability of reservoir bank slopes is directly related to disaster prevention and reduction and the safety of reservoir operation, deformation monitoring can provide early geological anomaly information, providing a scientific basis for predicting potential disasters and formulating prevention and control measures. Further, in order to more efficiently respond to sudden disasters, deformation monitoring and early warning processing can issue early warning signals in a timely manner by dynamically analyzing deformation characteristics and trends and combining with an early warning model, helping relevant departments to take intervention measures and minimizing disaster risks. Monitoring and early warning processing based on InSAR (Interferometric Synthetic Aperture Radar) data can make full use of the advantages of this technology in wide-area and high-precision deformation monitoring, overcoming the problems of limited spatial coverage and high labor costs of traditional surface observation means. Especially under complex terrain and extreme climate conditions, the InSAR technology can efficiently obtain the time series and spatial distribution information of slope deformation in a non-contact manner, providing scientific, efficient and reliable technical support for the long-term monitoring and accurate early warning of reservoir bank slope deformation.

[0003] The existing technology for monitoring and warning the deformation of reservoir bank slopes based on InSAR data usually obtains the surface deformation information of reservoir bank slopes through multi-temporal synthetic aperture radar (InSAR) images and uses differential InSAR technology (DInSAR) to extract accurate deformation data. First, by registering and interfering with radar images of different periods, the amount of deformation in the slope area is calculated. These deformation data can reflect the displacement trend and its spatial distribution of the slope. Next, using the time series analysis technology of deformation data, the deformation rate and change trend of the slope are evaluated. Combining external factors such as reservoir impoundment and climate change, it is identified whether there is a potential risk of landslide or collapse. To achieve real-time warning, the system usually sets a threshold. When the deformation exceeds the set safety critical value, the warning mechanism is automatically triggered to send a warning signal to relevant monitoring departments. At the same time, by combining the slope stability analysis model, the technology can further predict the trend of deformation development, evaluate the possibility of disaster occurrence, and then propose prevention and control countermeasures. In short, the monitoring and warning processing technology based on InSAR data can accurately predict and early warn the deformation evolution of reservoir bank slopes through high-frequency, long-term, and wide-range deformation monitoring, combined with intelligent analysis and warning models, thus providing strong support for disaster prevention, mitigation, and emergency response.

[0004] The existing technology has the following deficiencies:

[0005] In the case of a sharp drop in the reservoir water level within a short period of time, the reservoir bank slope originally supported by the buoyancy of the water will suddenly lose support, resulting in a rapid break of the stress balance inside the slope body. Instantaneous instability may occur in local areas, forming a rapid landslide. The deformation rate of this instantaneous instability is much higher than that of the normal slow deformation process. However, InSAR monitoring relies on the cumulative deformation analysis of multi-temporal images, and its data acquisition has a long time interval, making it difficult to continuously obtain deformation data in a short time, thus unable to capture the high-frequency dynamic changes of the slope after the sudden drop in water level. Therefore, the existing technology for monitoring and warning the deformation of reservoir bank slopes based on InSAR data cannot evaluate the risk of instantaneous instability of the slope in the case of a sharp drop in the reservoir water level within a short period of time, resulting in the system's difficulty in predicting the landslide trend of the slope body in a short time. This limitation may cause a warning lag, obtaining deformation data only after the landslide occurs, affecting the timeliness of disaster prevention response. At the same time, it misjudges the deformation trend, unable to accurately identify the acceleration stage before the landslide, increasing the deviation of risk assessment, and then missing the emergency response window, exacerbating the threat of the disaster to surrounding residents and infrastructure.

[0006] The above information disclosed in the background art section is only used to strengthen the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0007] The object of the present invention is to provide a method for monitoring, warning and processing the deformation of the reservoir bank slope based on InSAR data to solve the problems in the above-mentioned background technology.

[0008] To achieve the above object, the present invention provides the following technical solution: A method for monitoring, warning and processing the deformation of the reservoir bank slope based on InSAR data, specifically including the following steps:

[0009] In the reservoir bank slope area of the reservoir, the water level change information of the reservoir bank slope is obtained in real time through InSAR image data and water level monitoring data, and after obtaining, it is analyzed to judge whether the water level drop rate exceeds a preset threshold. If the water level drop rate exceeds the preset threshold, it is determined that the current situation is a sharp drop in the reservoir water level within a short time.

[0010] In the case of determining a sharp drop in the reservoir water level within a short time, the slope deformation characteristic information affected by the sudden drop in water level is obtained in real time, and after obtaining, it is analyzed to generate a water level sudden drop deformation response coefficient and an instantaneous instability trend index respectively.

[0011] An instantaneous instability risk assessment model is constructed for the generated water level sudden drop deformation response coefficient and instantaneous instability trend index to generate a slope instantaneous instability assessment coefficient.

[0012] Based on the slope instantaneous instability assessment coefficient, an assessment and analysis is carried out to evaluate the slope instantaneous instability risk in the case of a sharp drop in the reservoir water level within a short time, and it is divided into low risk, medium risk and high risk.

[0013] According to the assessment results, corresponding warning and processing measures are implemented for different risk levels.

[0014] Continuously monitor the deformation of the reservoir bank slope under the condition of water level drop, and after the warning and processing measures are executed, track its effect in real time. If the slope instantaneous instability assessment result shows that there is still a high risk, adjust the monitoring strategy and dynamically optimize the warning mechanism.

[0015] Preferably, in the case of determining a sharp drop in the reservoir water level within a short time, the slope deformation characteristic information affected by the sudden drop in water level is obtained in real time, and after obtaining, it is analyzed to generate a water level sudden drop deformation response coefficient and an instantaneous instability trend index respectively, specifically including the following steps:

[0016] In the case of determining a sharp drop in the reservoir water level within a short time, the slope deformation characteristic information affected by the sudden drop in water level is obtained in real time, and after obtaining, it is preprocessed.

[0017] Extract the water level sudden drop deformation characteristic information and slope instability evolution information from the preprocessed slope deformation characteristic information.

[0018] Analyze the extracted rapid drawdown deformation characteristic information and slope instability evolution information, and generate a rapid drawdown deformation response coefficient and an instantaneous instability trend index respectively.

[0019] Preferably, the acquisition logic of the rapid drawdown deformation response coefficient is as follows:

[0020] Extract the rapid drawdown deformation characteristic information from the preprocessed slope deformation characteristic information, specifically including the drawdown rate of the water level at different time points during a period of time during the reservoir water level decline, the surface displacement change rate of the slope in the horizontal distance, and the deformation acceleration of the slope surface, and represent them respectively by functions RW(t), GD(t), and VS(t) according to the time series, where t is the time point, RW(t) represents the drawdown rate of the water level at time point t during a period of time during the reservoir water level decline, GD(t) represents the surface displacement change rate of the slope in the horizontal distance at time point t during a period of time during the reservoir water level decline, VS(t) represents the deformation acceleration of the slope surface at time point t during a period of time during the reservoir water level decline, and the defined time period is [t1, t2];

[0021] Calculate the rapid drawdown deformation response coefficient, and the specific calculation formula is as follows:

[0022]

[0023] In the formula, CRD is the rapid drawdown deformation response coefficient.

[0024] Preferably, the acquisition logic of the instantaneous instability trend index is as follows:

[0025] Extract the slope instability evolution information from the preprocessed slope deformation characteristic information, specifically including the average displacement rate of the slope deep part, the stress change rate of the slope body, and the deformation rate acceleration of the slope surface at different time points during a period of time during the reservoir water level decline, and represent them respectively by functions VD(t), SS(t), and AV(t) according to the time series, where t is the time point, VD(t) represents the average displacement rate of the slope deep part at time point t during a period of time during the reservoir water level decline, SS(t) represents the stress change rate of the slope body at time point t during a period of time during the reservoir water level decline, AV(t) represents the deformation rate acceleration of the slope surface at time point t during a period of time during the reservoir water level decline, and the defined time period is [t1, t2];

[0026] Calculate the instantaneous instability trend index, and the specific calculation formula is as follows:

[0027]

[0028] In the formula, ITI is the instantaneous instability trend index.

[0029] Preferably, an instantaneous instability risk assessment model is constructed for the generated rapid drawdown deformation response coefficient CRD and instantaneous instability trend index ITI, and the instantaneous instability assessment coefficient of the slope is generated by weighted summation. The specific calculation formula is as follows:

[0030] SEC = ω1 * CRD + ω2 * ITI

[0031] In the formula, SEC is the instantaneous instability assessment coefficient of the slope, ω1 and ω2 are non-zero weight coefficients of the rapid drawdown deformation response coefficient CRD and the instantaneous instability trend index ITI respectively, and ω1 + ω2 = 1.

[0032] Preferably, a preset threshold interval [SEC min , SEC max of the instantaneous instability assessment coefficient of the slope is determined, and after determination, it is compared with the generated instantaneous instability assessment coefficient SEC of the slope. According to the comparison result, the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short time is evaluated and divided into low risk, medium risk and high risk. The specific comparison and analysis are as follows:

[0033] If SEC < SEC min , the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short time is low risk;

[0034] If SEC min ≤ SEC ≤ SEC max , the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short time is medium risk;

[0035] If SEC > SEC max , the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short time is high risk.

[0036] Preferably, according to the evaluation results, corresponding early warning treatment measures are implemented for different risk levels, specifically:

[0037] For the situation with a high-risk evaluation result, the specific early warning treatment measures are as follows: immediately activate the high-level landslide early warning mechanism, send high-priority warning information to the management department, emergency rescue units and surrounding residents; synchronously adjust the monitoring mode to increase the acquisition frequency of InSAR data; take temporary slope reinforcement measures; at the same time, adjust the reservoir operation strategy to reduce the risk of further inducing instability, and continuously monitor the slope state;

[0038] For the case where the assessment result is medium risk, the specific early warning treatment measures are as follows: Trigger the conventional slope early warning mechanism, send medium-priority early warning information to the monitoring unit and the reservoir management party, and remind them to strengthen monitoring; Dynamically adjust the monitoring frequency, increase the update frequency of InSAR images, and increase the monitoring intensity of surface strain sensors; Combine the reservoir operation plan to optimize the water level decline rate; Adjust the threshold update strategy of the early warning system;

[0039] For the case where the assessment result is low risk, the specific early warning treatment measures are as follows: Keep the current monitoring strategy unchanged, and maintain the regular monitoring frequency of InSAR image updates and surface strain monitoring; Continuously track the water level decline rate and the slope deformation trend to ensure the stability of the assessment result.

[0040] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0041] 1. By constructing the water level sudden drop deformation response coefficient and the instantaneous instability trend index, the present invention can accurately quantify the impact of the water level sudden drop on slope deformation, and combined with the slope instantaneous instability assessment coefficient, it realizes the dynamic monitoring and risk grading of slope deformation in a short time. Compared with the traditional InSAR deformation monitoring technology that relies on long-term image cumulative analysis, the present invention can obtain the slope deformation characteristics within a short time (such as within several hours) after the water level sudden drop, and calculate the instability trend in real time through mathematical modeling, providing high-timeliness and high-precision technical support for the safety assessment of the reservoir bank slope.

[0042] 2. The present invention realizes the full-process intelligence from data collection, mathematical modeling, risk assessment to early warning treatment. Through multi-source data fusion (InSAR images, GNSS monitoring points, tilt sensors, acceleration sensors, and water level monitoring data) combined with advanced mathematical modeling (exponential operation, logarithmic operation, integral calculation), the present invention improves the analysis ability of slope deformation and can accurately identify the trend of the slope entering the accelerating stage of instability during the water level sudden drop. In addition, by constructing a risk grading mechanism through the slope instantaneous instability assessment coefficient, the present invention can divide the slope instantaneous instability risk into low risk, medium risk, and high risk, and match different levels of early warning treatment measures to ensure that the management department can accurately implement risk intervention according to real-time data, improving the scientificity and reliability of disaster early warning.

[0043] 3. Another technical advantage of the present invention lies in the adaptive dynamic monitoring and optimization warning mechanism, which ensures that during the continuous change of the reservoir water level, the monitoring and warning strategies can be intelligently adjusted. After the warning is executed, the system continuously monitors the slope deformation situation and tracks the effect of the warning measures in real time. If the instantaneous instability assessment result of the slope is still in a high-risk state, the system will dynamically optimize the monitoring strategy, including increasing the GNSS and InSAR data acquisition frequencies, optimizing the data fusion algorithm, adjusting the reservoir operation strategy, etc., so as to ensure the accuracy and effectiveness of slope warning. Compared with the traditional monitoring methods, which have problems such as warning lag, risk misjudgment, and insufficient emergency response, the present invention can provide an early warning before the disaster occurs and has the ability to actively respond to different risk levels, providing an intelligent, efficient, and real-time solution for the safety management of the reservoir bank slope. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0045] Figure 1 It is a schematic flowchart of a method for monitoring and warning the deformation of a reservoir bank slope based on InSAR data of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] Now, the exemplary embodiments will be described more comprehensively with reference to the drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as being limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art.

[0047] The present invention provides a method for monitoring and warning the deformation of a reservoir bank slope based on InSAR data as shown in Figure 1 the following, which specifically includes the following steps:

[0048] In the reservoir bank slope area, the water level change information of the reservoir bank slope is obtained in real time through InSAR image data and water level monitoring data, and after obtaining, it is analyzed to determine whether the water level decline rate exceeds a preset threshold. If the water level decline rate exceeds the preset threshold, it is determined that the current situation is a sharp decline in the reservoir water level within a short period of time;

[0049] The water level change information of the reservoir bank slope can be obtained in real time by combining satellite InSAR image data with water level monitoring data. The InSAR image data can be regularly acquired by radar satellites (such as Sentinel-1, TerraSAR-X), and through image preprocessing, including steps such as geometric correction, coherence analysis, phase unwrapping, and time series processing, high-precision surface deformation data can be obtained. At the same time, the water level monitoring data can be collected in real time through water level sensors, buoy measurement systems, radar depth sounders, or GNSS stations in the reservoir, and transmitted to the data processing system through wireless communication to ensure the real-time and continuity of the data. To improve the spatial accuracy of the data, the pixel-level water level change information of the InSAR image can be combined, and the point data of the water level sensor can be extended to the entire reservoir bank slope area through a data fusion algorithm to form a complete water level change field.

[0050] After obtaining the data, the time series analysis method can be used to determine whether the water level decline rate exceeds the preset threshold. First, the difference calculation is performed on the water level data at consecutive times to obtain the water level change rate at different time points, and the sliding window method is used to smooth the data to reduce the interference of instantaneous fluctuations. Subsequently, the system will compare the calculated water level decline rate with the preset threshold. If the decline rate is greater than the threshold, it is considered that the water level decline rate has reached the critical state. The preset threshold can be determined by analyzing historical water level data, statistical analysis methods, machine learning models, etc., specifically including (1) analyzing historical reservoir operation data and calculating the normal rate range of water level decline over the years; (2) based on the geological characteristics of the reservoir bank slope, combining with the slope stability model to determine the influence of slope stress changes corresponding to different water level decline rates; (3) using an anomaly detection model based on machine learning to train the relationship between the water level decline rate and the slope deformation trend, and automatically determine the optimal threshold range.

[0051] After determining that the water level decline rate exceeds the preset threshold, the system will automatically determine that the current situation is a sharp decline in the reservoir water level in a short time. The specific implementation methods include: First, the system will trigger the dynamic monitoring mode, increase the data collection frequency, improve the acquisition density of InSAR images, and adjust the data update interval of the water level sensor to record the water level decline trend in real time. Secondly, the system will input the change curve of the water level decline rate into the sudden water level decline trend analysis model to calculate the change trend of the water level decline rate. If the system detects that the water level decline rate continues to be higher than the threshold and the water level change trend in a short time conforms to the historical sudden decline mode, it will further confirm that the current situation is a sharp decline in the reservoir water level in a short time. In addition, the system can use prediction algorithms based on Bayesian inference or time series deep learning models to fit the future water level change trend to determine whether it conforms to the characteristics of a sharp decline in a short time.

[0052] The purpose of doing this is to promptly identify the sharp decline in the reservoir water level within a short period of time, thereby predicting and evaluating in advance the instantaneous instability risk of the reservoir bank slope, and improving the accuracy of monitoring and early warning. The existing InSAR monitoring methods mainly rely on the analysis of long-term deformation accumulation and it is difficult to continuously obtain high-frequency deformation data of the slope within a short period of time. Therefore, it is impossible to effectively identify the precursors of instantaneous instability. This solution enables the system to detect risks at the initial stage of a sudden water level drop through software methods such as real-time water level monitoring, dynamic threshold comparison, and trend analysis. Furthermore, it jointly evaluates the instantaneous instability trend of the slope and takes more accurate early warning measures to reduce the occurrence probability of landslide disasters and improve the efficiency of disaster prevention and mitigation.

[0053] When it is determined that the reservoir water level drops sharply within a short period of time, the deformation characteristic information of the slope affected by the sudden water level drop is obtained in real time and analyzed after acquisition to generate a deformation response coefficient for sudden water level drop and an instantaneous instability trend index respectively.

[0054] In this embodiment, when it is determined that the reservoir water level drops sharply within a short period of time, the deformation characteristic information of the slope affected by the sudden water level drop is obtained in real time and analyzed after acquisition to generate a deformation response coefficient for sudden water level drop and an instantaneous instability trend index respectively, which specifically includes the following steps:

[0055] When it is determined that the reservoir water level drops sharply within a short period of time, the deformation characteristic information of the slope affected by the sudden water level drop is obtained in real time and preprocessed after acquisition.

[0056] The real-time acquisition of the deformation characteristic information of the slope affected by the sudden water level drop can be achieved through the fusion of multi-source monitoring data, which mainly includes the comprehensive acquisition and processing of InSAR image data, water level monitoring data, and surface deformation sensing data. First, InSAR image data can be obtained through regularly acquired satellite radar images (such as Sentinel-1, TerraSAR-X), and the slope deformation information can be extracted using time series interferometry technology. Phase unwrapping and accuracy correction are performed through image analysis algorithms to ensure the spatial continuity of the data. Second, the water level monitoring data can be used to obtain the water level changes in real time through ultrasonic water level gauges, radar water level gauges, or buoy-type water level monitors deployed in the reservoir, and the data is transmitted to the monitoring system through wireless communication technology to calculate the water level drop rate and identify sudden water level drop events. In addition, the surface deformation sensing data is collected in real time by GNSS monitoring points, tilt sensors, and crack displacement gauges deployed on the reservoir bank slope. The displacement, tilt, and shear deformation information obtained by each sensor is spatially matched through a data fusion algorithm to ensure the time synchronization and accuracy of the data. The system starts a high-frequency monitoring mode after detecting that the water level drop rate exceeds the set threshold through an automatic trigger mechanism, increases the acquisition frequency of InSAR data, and simultaneously improves the data acquisition frequency of the sensors to ensure real-time performance and accuracy.

[0057] The purpose of preprocessing is to improve the stability of data, remove noise interference, unify the spatio-temporal resolution, and enhance the comparability of different data sources to ensure the accuracy and reliability of the calculated slope deformation characteristic information. First, data alignment and interpolation processing: Since the acquisition frequencies and spatial resolutions of InSAR image data, water level monitoring data, and sensor data are different, it is necessary to perform time alignment and spatial interpolation on the data to ensure that the data at the same time point can correspond to the same slope area. Second, noise filtering and outlier removal: InSAR data is affected by atmospheric delay, coherence change, etc., and it is necessary to remove atmospheric noise through coherence analysis and time series filtering algorithms (such as SBAS, PS-InSAR); abnormal jumps may occur in water level sensing data due to sensor errors, and sliding mean filtering and Kalman filtering can be used to smooth the data; abnormal data points in sensor displacement data can be removed through median filtering. Finally, data fusion and format standardization: The data formats from different sources may be inconsistent, and it is necessary to unify them into a standardized format and construct a spatial data fusion model, and use the weight weighting algorithm to optimize the fusion accuracy of different data sources to ensure that the extracted slope deformation characteristic information can reflect the real impact of sudden water level drop on slope deformation. All preprocessing processes are automatically executed by the software module of the monitoring system, and the preprocessing strategy can be dynamically adjusted according to the real-time data quality to improve the accuracy of the final calculation results.

[0058] Extract the sudden water level drop deformation characteristic information and slope instability evolution information from the preprocessed slope deformation characteristic information, which are respectively used to characterize the impact of sudden water level drop on the surface deformation of the slope and the instability trend of the deep structure of the slope;

[0059] The water level sudden drop deformation feature information and slope instability evolution information can be extracted from the preprocessed slope deformation feature information through data classification and feature extraction algorithms. First, the system needs to divide the labels of the preprocessed data set, that is, divide the data into water level sudden drop related data and slope instability evolution related data based on the data source, physical meaning, and calculation target. Specifically, data stratification indexing can be used to classify the data according to the surface deformation characteristics and deep instability characteristics driven by the water level sudden drop. When extracting the water level sudden drop deformation feature information, the system will screen data such as the water level drop rate, surface displacement gradient, and slope surface deformation rate, and apply time series analysis algorithms (such as moving window mean, weighted time series model) to calculate the dynamic response of the slope surface deformation. At the same time, when extracting the slope instability evolution information, the system will select data such as the deep displacement rate, shear strain change rate, and deformation rate acceleration, and extract the instability characteristics of the deep structure of the slope body through deformation trend analysis (such as curve fitting, second derivative calculation). In addition, to improve the accuracy of data classification, principal component analysis (PCA) or clustering algorithms (such as K-means) can be used to automatically determine which data variables contribute the most to the water level sudden drop deformation and slope instability evolution, and perform optimal grouping. After the data is extracted, the system will perform normalization processing to ensure that the data units and scales are consistent, and finally store the water level sudden drop deformation feature information and slope instability evolution information in different logical partitions of the database for subsequent calculation of the water level sudden drop deformation response coefficient and instantaneous instability trend index.

[0060] Analyze the extracted water level sudden drop deformation feature information and slope instability evolution information to generate the water level sudden drop deformation response coefficient and instantaneous instability trend index respectively.

[0061] In this embodiment, the acquisition logic of the water level sudden drop deformation response coefficient is as follows:

[0062] Extract the water level sudden drop deformation feature information from the preprocessed slope deformation feature information, specifically including the water level drop rate at different time points within a period during the reservoir water level drop process, the surface displacement change rate of the slope in the horizontal distance, and the deformation acceleration of the slope surface, and represent them respectively by functions RW(t), GD(t), and VS(t) according to the time series, where t is the time point, RW(t) represents the water level drop rate at time point t within a period during the reservoir water level drop process, GD(t) represents the surface displacement change rate of the slope in the horizontal distance at time point t within a period during the reservoir water level drop process, VS(t) represents the deformation acceleration of the slope surface at time point t within a period during the reservoir water level drop process, and the defined time period is [t1, t2];

[0063] The real-time acquisition of the water level drop rate, the surface displacement change rate of the slope in the horizontal distance, and the deformation acceleration of the slope surface can be achieved through a multi-source data fusion and real-time monitoring system, and the accuracy and timeliness of the data are ensured through data preprocessing and fusion algorithms. First, the water level drop rate can be measured in real time by water level monitoring sensors (such as ultrasonic water level gauges, radar water level gauges, buoy-type water level monitors) and transmitted to the data processing system through a wireless communication network (such as LoRa, 5G, or satellite communication). The system calculates the water level drop rate based on the continuous time series and filters out abnormal data to ensure data stability. Second, the surface displacement change rate of the slope in the horizontal distance can be calculated by fusing InSAR image data and GNSS monitoring point data. Among them, the InSAR technology (such as Sentinel-1 data combined with SBAS or PS-InSAR algorithms) can obtain large-scale surface deformation information, while the GNSS monitoring points (deployed on the typical slip zones of the reservoir bank slope) can provide high-precision displacement change data. The data fusion adopts spatio-temporal matching and weighted interpolation methods to ensure that the areal data of the InSAR image and the point data of the GNSS are calculated on the same time basis, so as to obtain the surface displacement gradient of the slope at different time points. Finally, the deformation acceleration of the slope surface can be obtained in real time by high-frequency dynamic monitoring devices (such as fiber Bragg grating sensors, MEMS acceleration sensors, triaxial inclinometers). These devices are installed on the surface layer of the slope and record the slope vibration, tilt, and acceleration changes at a high sampling rate (such as 1 - 10 Hz per second). The data is wirelessly transmitted through the slope monitoring gateway. The system uses the short-time Fourier transform (STFT) or wavelet transform to analyze the time series data, extracts the deformation acceleration at different time points, removes background noise and interference signals, and improves the monitoring accuracy. After all the data enters the database, the software system unifies these data to the same time coordinate system through data alignment, filtering, and time series analysis for subsequent calculation of the deformation response coefficient of the sudden water level drop.

[0064] Calculate the deformation response coefficient of the sudden water level drop. The specific calculation formula is as follows:

[0065]

[0066] In the formula, CRD is the deformation response coefficient of the sudden water level drop.

[0067] The design of the calculation formula for the deformation response coefficient CRD of the sudden water level drop is based on the non-linear influence of the water level drop on the surface deformation of the slope, and a mathematical method is used to reasonably quantify this relationship to ensure the accuracy and physical meaning of the evaluation results. First, the integration operation accumulates and calculates the data within the time period [t1, t2] to ensure that the calculation result can reflect the overall trend of the slope deformation during the entire sudden water level drop process, rather than just the instantaneous value at a certain moment. The denominator For normalization calculation, it ensures that the calculation results remain relatively stable regardless of the length of the selected time period, avoiding numerical deviations caused by different sizes of time windows.

[0068] Inside the integral, first, GD(t) (ground surface displacement change rate) is used. It describes the deformation change amount of the slope in the horizontal distance and directly reflects the slope strain state caused by the sudden drop of the water level. This term is used as the main variable to ensure that the calculation focuses on the deformation degree of the slope. Secondly, e RW ( t ) amplifies the influence of the water level drop rate RW(t) on the deformation through exponential operation. Because the faster the water level drops, the deformation response of the slope usually increases non-linearly. Therefore, the exponential transformation helps to more realistically describe this phenomenon. Finally, ln(1 + VS(t)) processes the slope deformation acceleration VS(t) using logarithmic operation, avoiding the instability of the calculation results caused by extreme values of the deformation acceleration and ensuring a significant influence even at low values. The introduction of logarithmic operation can emphasize the relative change rate of the deformation acceleration rather than the absolute value, thus enhancing the stability and adaptability of the calculation.

[0069] In summary, each part of this calculation formula has a clear physical meaning. The integral calculation is used to obtain the overall trend over the entire time period, the exponential operation amplifies the influence of the sudden water level drop rate, and the logarithmic operation smooths the effect of the deformation acceleration, ultimately achieving an accurate characterization of the deformation response to the sudden water level drop.

[0070] The magnitude of the deformation response coefficient CRD to the sudden water level drop directly affects the assessment of the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short period, because it quantifies the response degree of the sudden water level drop to the slope deformation. When the CRD value is large, it indicates that during the sudden water level drop, the ground surface deformation of the slope is significant, the displacement gradient is high, and the slope deformation rate changes violently with the water level drop. This situation usually means that the slope stress is significantly adjusted, which may lead to an increase in strain accumulation and then trigger instantaneous instability. Especially when the sudden water level drop rate is fast, the exponential term causes the CRD value to rise sharply, indicating that the driving force of the sudden water level drop on the slope deformation is significantly enhanced. At this time, the non-linear effect of the slope deformation is intensified and the landslide risk increases. On the contrary, when the CRD value is small, it indicates that the influence of the water level drop on the slope deformation is small, the slope as a whole maintains good stability, and it is not easy to become unstable in a short period.

[0071] In this embodiment, the acquisition logic of the instantaneous instability trend index is as follows:

[0072] Extract the slope instability evolution information from the preprocessed slope deformation characteristic information, specifically including the average displacement rate of the deep part of the slope, the stress change rate of the slope body, and the deformation rate acceleration of the slope surface at different time points during a period of time during the reservoir water level decline. And they are respectively represented by functions VD(t), SS(t), and AV(t) according to the time series, where t is the time point. VD(t) represents the average displacement rate of the deep part of the slope at time t during a period of time during the reservoir water level decline, SS(t) represents the stress change rate of the slope body at time t during a period of time during the reservoir water level decline, and AV(t) represents the deformation rate acceleration of the slope surface at time t during a period of time during the reservoir water level decline. The defined time period is [t1, t2].

[0073] The real-time acquisition of the average displacement rate of the deep part of the slope, the stress change rate of the slope body, and the deformation rate acceleration of the slope surface can be achieved through a distributed monitoring and sensing system and data fusion algorithms, and a software system is used for data processing and analysis to ensure the accuracy and timeliness of the data. First, the average displacement rate of the deep part of the slope can be obtained through buried GNSS monitoring points, borehole strain gauges, and distributed fiber optic sensors (DFOS). The GNSS monitoring points are installed at key parts of the internal structure of the slope, and the real-time kinematic positioning (RTK-GNSS) is used to calculate the displacement rate of the deep rock mass. The borehole strain gauges can measure the displacement changes of the rock mass at different depths, and the average rate is calculated by the time series curve fitting method. The DFOS measures the deformation of the deep geotechnical layer by arranging fiber Bragg gratings (FBG) along the slope body. The software system calculates the average displacement rate of the entire deep part of the slope through data fusion algorithms (such as weighted average or Bayesian estimation). Second, the stress change rate of the slope body can be measured through embedded strain sensors, earth pressure gauges, and microseismic monitoring systems. The embedded strain sensors are installed in the key fault zones of the slope body, and the local stress change rate is calculated through the strain-stress conversion model. The earth pressure gauges are arranged near the slip surface to monitor the stress changes of the slip zone in real time. After the data is corrected by filtering and interpolation algorithms, the stress change rate is calculated. The microseismic monitoring system identifies the micro-fracture events inside the slope through vibration spectrum analysis and estimates the overall stress change rate of the slope in combination with the historical stress evolution model. Finally, the deformation rate acceleration of the slope surface can be calculated through triaxial inclinometers, MEMS accelerometers, and InSAR image data. The triaxial inclinometers can measure the angle change rate of the slope surface and calculate the deformation acceleration through the second derivative. The MEMS accelerometers are installed on the surface of the slope body to record the vibration data of the slope surface in real time. The software system uses the short-time Fourier transform (STFT) or wavelet transform to extract the deformation acceleration signal. The InSAR image data calculates the surface deformation rate through multi-temporal interferometry techniques (such as SBAS or PS-InSAR) and calculates the acceleration through the second-order time series difference. After all the data is processed by spatio-temporal matching, noise filtering, and data fusion, the system can uniformly calculate these key parameters in the same time coordinate system to ensure the accuracy and real-time of the calculation of the instantaneous instability trend index.

[0074] Calculate the instantaneous instability trend index, and the specific calculation formula is as follows:

[0075]

[0076] In the formula, ITI is the instantaneous instability trend index.

[0077] The calculation formula of the instantaneous instability trend index ITI is based on the physical mechanism of slope instability evolution. Integral operation is used to comprehensively quantify the deep displacement rate, slope stress change rate, and deformation rate acceleration of the slope during the water level drop process to evaluate the instability trend of the slope. First, the integral operation is used to cumulatively calculate the deformation trend within the calculation time period [t1, t2] to ensure that the calculation result can reflect the overall change in slope stability during the entire water level drop process, rather than just a certain instantaneous state. At the same time, the denominator is normalized so that the calculation result will not cause numerical deviation due to different time windows, improving the robustness of the calculation.

[0078] Inside the integral, first, the average displacement rate VD(t) of the deep part of the slope is used. This term reflects the displacement trend of the deep structure of the slope. Since slope instability usually starts from deep shear failure, VD(t) is the core index of the instability trend. Second, the slope stress change rate SS(t) is logarithmically transformed as ln(1 + SS(t)) because the stress change often has a large dynamic range. Directly using the original value may cause the calculated value to fluctuate violently, while the logarithmic operation can map a large range of data to a relatively stable numerical space and ensure that the influence at low values is still significant. Finally, e -AV ( t ) modulates the acceleration of the deformation rate AV(t) through exponential decay operation. The greater the deformation acceleration, the closer the exponential term e -AV ( t ) is to 0, indicating that the slope is rapidly entering the unstable state. At this time, the ITI value rises, strengthening the influence of the deformation acceleration on the instability trend. In summary, this calculation formula uses integral operation to ensure the integrity of trend quantification, exponential operation to emphasize the non-linear amplification effect of the deformation rate acceleration on instability, and logarithmic operation to balance the influence of the stress change rate, so that the calculation result can accurately characterize the instability trend of the slope.

[0079] The magnitude of the instantaneous instability trend index ITI directly reflects the instability evolution trend of the slope under the condition of a sharp drop in the reservoir water level within a short period of time, and can be used to quantify the possibility of the slope entering the instantaneous instability state. When the ITI value is large, it indicates that the displacement rate in the deep part of the slope is high, the stress change rate of the slope body increases rapidly, and the acceleration of the deformation rate is significant. This shows that the deformation of the slope accelerates within a short period of time, the instability trend intensifies, and the possibility of entering the landslide state increases significantly, belonging to a high-risk state. On the contrary, when the ITI value is small, it means that the deformation in the deep part of the slope is small, the stress change of the slope body is slow, and the acceleration of the deformation rate is low, indicating that the slope structure is still in a relatively stable state. At this time, no additional intervention measures are required, and only routine monitoring needs to be maintained. Therefore, the magnitude of the ITI value is positively correlated with the instantaneous instability risk of the slope under the condition of a sudden drop in the reservoir water level. By quantifying the comprehensive influence of the deep displacement of the slope, the stress change of the slope body, and the acceleration of the deformation rate, it provides an accurate basis for risk assessment, thereby guiding corresponding early warning and intervention decisions.

[0080] Construct an instantaneous instability risk assessment model for the generated deformation response coefficient of the sudden drop in water level and the instantaneous instability trend index, and generate the instantaneous instability assessment coefficient of the slope;

[0081] In this embodiment, an instantaneous instability risk assessment model is constructed for the generated deformation response coefficient CRD of the sudden drop in water level and the instantaneous instability trend index ITI, and the instantaneous instability assessment coefficient of the slope is generated by weighted summation. The specific calculation formula is as follows:

[0082] SEC = ω1 * CRD + ω2 * ITI

[0083] In the formula, SEC is the instantaneous instability assessment coefficient of the slope, ω1 and ω2 are the non-zero weight coefficients of the deformation response coefficient CRD of the sudden drop in water level and the instantaneous instability trend index ITI respectively, and ω1 + ω2 = 1.

[0084] The calculation of the instantaneous instability assessment coefficient SEC of the slope is based on the weighted summation of the deformation response coefficient CRD of the sudden drop in water level and the instantaneous instability trend index ITI to comprehensively evaluate the risk degree of the slope occurring instantaneous instability when the reservoir water level drops sharply within a short period of time. This calculation formula reflects the interaction between the deformation response of the slope within a short period of time and the instability trend of the deep structure, ensuring that the evaluation result can not only reflect the direct impact of the sudden drop in water level on the surface deformation, but also reflect the stability evolution of the deep structure of the slope. The weight coefficients ω1 and ω2 in the formula are used to balance the contributions of these two key parameters to the overall evaluation, and satisfy the constraint condition ω1 + ω2 = 1 to ensure the normalized calculation of the evaluation coefficient, making SEC have a unified physical meaning.

[0085] The weight coefficients ω1 and ω2 represent the relative importance of the water level sudden drawdown deformation response and the instantaneous instability trend in the final risk assessment, which can be determined according to the historical monitoring data of the slope or machine learning optimization algorithms. In areas where the slope is sensitive to the height of the water level sudden drawdown, ω1 is usually larger because the instability of the slope is mainly driven by the water level change and the role of surface deformation is more significant; while in slopes controlled by deep structures for instability (such as areas with slip surfaces), ω2 will be larger to highlight the influence of the deep displacement rate and stress change. In addition, the weights can be adjusted through historical data regression or Bayesian optimization methods to ensure that the evaluation model can adapt to different slope geological conditions and improve the accuracy of the evaluation. When the SEC exceeds the high-risk threshold, it means that the slope is in a high instability risk state, and warning and emergency response measures should be taken immediately. When the SEC is in the low-risk range, it indicates that the slope remains stable in the short term and no additional intervention is required.

[0086] Based on the instantaneous instability assessment coefficient of the slope, the assessment and analysis are carried out to evaluate the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short time, and it is divided into low risk, medium risk and high risk;

[0087] In this embodiment, the threshold interval [SEC min , SEC max of the pre-set instantaneous instability assessment coefficient of the slope is determined, and after determination, it is compared with the generated instantaneous instability assessment coefficient SEC of the slope. According to the comparison result, the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short time is evaluated and divided into low risk, medium risk and high risk. The specific comparison and analysis are as follows:

[0088] If SEC < SEC min , the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short time is low risk;

[0089] This situation means that the sharp drop in the reservoir water level in a short time has little impact on the slope, and the slope as a whole remains stable. At this time, the surface deformation of the slope is relatively slow, and the deep displacement rate and stress change rate do not reach a significant level, indicating that the slope has not entered the instability evolution stage. In this case, the possibility of landslide or local instability is extremely low. Therefore, no additional warning intervention is required, and only data collection and analysis need to be carried out according to the regular monitoring plan to ensure that the slope state is still within the safe range. If the future water level drop rate increases or the slope deformation trend intensifies, the monitoring strategy can be adjusted in a timely manner to prevent potential risks.

[0090] If SEC min ≤ SEC ≤ SEC max , the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level in a short time is medium risk;

[0091] This situation means that the impact of the sudden water level drop on the slope deformation has become apparent, and the slope deformation rate has increased, but it has not reached the critical instability state. At this time, the surface displacement gradient, the deep displacement rate, or the change rate of slope stress may show a stagewise increase, and the slope is in a potentially unstable state. If the water level continues to drop in the future or the external environment changes (such as rainfall, earthquake, etc.), the instability risk may increase further. In this case, it is necessary to strengthen the monitoring frequency, such as increasing the sampling rate of GNSS monitoring data, shortening the acquisition period of InSAR data, or introducing monitoring means with higher resolution, so as to timely grasp the change trend of the slope. At the same time, the reservoir operation strategy can be adjusted appropriately according to the risk level to reduce the possibility of further instability.

[0092] If SEC > SEC max , the instantaneous instability risk of the slope under the condition of a sharp drop in the reservoir water level within a short time is a high risk.

[0093] This situation means that the slope has entered the critical instability state, and the risk of geological disasters such as landslides and collapses occurring within a short time is extremely high. At this time, the impact of the sudden water level drop on the slope has caused obvious non-uniform deformation, the surface displacement has intensified, the deep displacement rate has risen rapidly, and the change rate of shear strain shows a non-linear growth trend, indicating that the slope structure may be undergoing irreversible changes. In this case, the emergency warning mechanism should be activated immediately, including sending landslide warning signals to relevant management departments and personnel, formulating emergency evacuation plans, and taking engineering reinforcement measures (such as retaining, anti-seepage, anchoring, etc.) if necessary to reduce the possibility of slope instability. At the same time, the reservoir operation plan should also be adjusted to avoid the further aggravation of the slope instability state caused by the sudden water level drop, so as to minimize the risk and loss of disasters.

[0094] The threshold interval of the preset instantaneous slope instability assessment coefficient can be determined by combining historical data analysis, machine learning modeling, and slope stability theoretical calculations, and an automated calculation and optimization adjustment can be performed using a software system to adapt to different geological environments and water level changes. First, based on historical data analysis, the system can retrieve long-term reservoir monitoring data, including InSAR deformation data during the water level decline, GNSS displacement rates, stress change rates recorded by strain sensors, and historical landslide events. Statistical analysis methods (such as quantile analysis, K-means clustering) are used to classify the SEC values that have experienced instability historically, and initial threshold intervals for low risk, medium risk, and high risk are set accordingly. Second, based on machine learning modeling, the system can adopt supervised learning methods (such as random forest, support vector machine SVM, or XGBoost), input historical monitoring data and the corresponding instability risk levels, learn the mapping relationship between the SEC value and the actual instability event under sudden water level drops through training the model, and then use new data for real-time prediction to automatically adjust the threshold interval to dynamically adapt to different water level decline rates and slope geological conditions. Finally, based on slope stability theoretical calculations, the system can call numerical simulation methods such as finite element analysis (FEM) and limit equilibrium analysis (LEM), calculate the stability coefficient FS of the slope based on the pore water pressure changes, effective stress adjustments, and shear slip trends caused by sudden water level drops, and perform fitting in combination with the SEC value to match the threshold interval with the physical model. By integrating the above methods, the system can regularly update the threshold settings, automatically adjust the threshold interval when the water level decline rate or environmental conditions change, ensure the scientificity and reliability of the assessment, and at the same time improve the accuracy and adaptability of the early warning.

[0095] According to the evaluation results, corresponding early warning treatment measures are implemented for different risk levels;

[0096] In this embodiment, according to the evaluation results, corresponding early warning treatment measures are implemented for different risk levels, specifically:

[0097] For the case where the evaluation result is high risk, the specific early warning treatment measures are: immediately activate the high-level landslide early warning mechanism, send high-priority early warning information to the management department, emergency rescue units, and surrounding residents; synchronously adjust the monitoring mode, increase the acquisition frequency of InSAR data, and increase the real-time monitoring density of tilt sensors and seismic sensors; take temporary slope reinforcement measures, including but not limited to slope surface support, anchoring reinforcement, or slope cutting and load reduction; at the same time, adjust the reservoir operation strategy, reduce the water level decline rate or suspend water discharge to reduce the risk of further inducing instability, and continuously monitor the slope state. If the risk intensifies, activate the emergency evacuation plan;

[0098] For the situation where the evaluation result is high risk, the implementation method of the early warning treatment measures is as follows: When the instantaneous slope instability evaluation coefficient exceeds the high-risk threshold, the software system automatically triggers a high-level landslide early warning mechanism, and sends high-priority early warning information to the reservoir management department, emergency response units and surrounding residents through the slope monitoring platform. The information content includes the probability of landslide occurrence, the slope deformation trend and emergency evacuation suggestions. To ensure the early warning accuracy, the system dynamically adjusts the monitoring mode, immediately increases the data acquisition frequency of GNSS monitoring points (such as increasing from once every 10 minutes to once every 1 minute), calls high-resolution InSAR image data (such as the emergency programming mode of SAR satellites), and performs high-frequency sampling on the real-time data of tilt sensors and seismic sensors. The system uses data fusion algorithms (such as Kalman filtering, Bayesian inference) to automatically eliminate abnormal data and improve the early warning accuracy. At the same time, the software system is linked with the reservoir operation management platform, invokes the water level regulation algorithm, calculates the optimal water level decline rate, and sends adjustment suggestions to the dispatching center to control the water release rate to reduce the adverse impact of sudden water level drop on the slope. The purpose of doing this is to slow down the slope instability process as much as possible in the high-risk state, buy time for emergency response, and reduce the possibility of disasters.

[0099] For the situation where the evaluation result is medium risk, the specific early warning treatment measures are as follows: Trigger the conventional slope early warning mechanism, send medium-priority early warning information to the monitoring unit and the reservoir management party to remind them to strengthen monitoring; Dynamically adjust the monitoring frequency, increase the update frequency of InSAR images, and increase the monitoring intensity of surface strain sensors; Combine the reservoir operation plan to optimize the water level decline rate to avoid further deterioration of slope stability caused by continued decline in a short period of time; Adjust the threshold update strategy of the early warning system. If subsequent monitoring data shows an increasing instability trend, upgrade to the high-risk level and execute the corresponding emergency response measures;

[0100] For the situation where the assessment result is medium risk, the implementation method of the early warning treatment measures is as follows: When the SEC is in the medium-risk range, the system automatically triggers the regular slope early warning mechanism and sends medium-priority early warning information to the monitoring unit and the reservoir management party, including the current deformation trend of the slope, the possibility of instability, and the monitoring data analysis results. To strengthen the monitoring intensity, the system adopts an adaptive data sampling strategy, such as automatically adjusting the sampling interval of GNSS monitoring points (shortening from 30 minutes per time to 10 minutes per time), increasing the update frequency of InSAR images (such as adjusting from 6 days per time to 3 days per time), and performing real-time processing on the data of strain sensors on the slope surface to enhance the timeliness of the data. The system can also apply machine learning models (such as support vector machine SVM or time series LSTM network) to predict the deformation trend of the slope in the next 24-48 hours and optimize the reservoir water level regulation strategy accordingly to avoid continuing to accelerate water discharge when the slope deformation intensifies. The purpose of doing this is to prevent the risk from further intensifying by optimizing the monitoring and regulation measures when the instability trend has not reached the critical state, and at the same time ensure the continuous reliability of the slope monitoring data.

[0101] For the situation where the assessment result is low risk, the specific early warning treatment measures are as follows: Keep the current monitoring strategy unchanged, and maintain the regular monitoring frequency of InSAR image update and surface strain monitoring; Continuously track the water level drop rate and the slope deformation trend to ensure the stability of the assessment result. If the subsequent monitoring data shows that the slope deformation trend increases and the instantaneous instability assessment coefficient of the slope approaches the medium-risk level, adjust the monitoring strategy, increase the monitoring frequency, and optimize the data processing model to identify potential risks in advance.

[0102] For the situation where the assessment result is low risk, the implementation method of the early warning treatment measures is as follows: When the SEC is lower than the low-risk threshold, the system maintains the regular monitoring strategy and continuously tracks the water level change and the slope deformation trend. The software system can call time series trend analysis algorithms (such as moving average filtering, exponential weighted smoothing) to calculate the SEC change trend in the next period of time. If the trend is stable or decreasing, maintain the standard sampling frequency of the current GNSS monitoring, InSAR image update, and surface strain monitoring (such as 60 minutes per time for GNSS monitoring points, 6-12 days per time for InSAR images). At the same time, the software system continuously records the SEC data and judges whether it is necessary to update the SEC preset threshold through an adaptive threshold adjustment algorithm (such as dynamic quantile analysis) to adapt to environmental changes. If the system detects that the SEC is gradually approaching the medium-risk range, it automatically increases the monitoring frequency and optimizes the data processing model (such as increasing the weight of historical data to improve the prediction accuracy). The purpose of doing this is to avoid over-regulation when the slope deformation has little impact, reduce unnecessary consumption of monitoring resources, and at the same time ensure that the system can take actions in advance when the SEC starts to approach the critical value, improving the sensitivity and accuracy of the early warning.

[0103] Continuously monitor the deformation of the reservoir bank slope under the condition of water level decline, and track its effect in real time after the early warning treatment measures are implemented. If the instantaneous instability assessment result of the slope still shows high risk, adjust the monitoring strategy and dynamically optimize the early warning mechanism.

[0104] In order to continuously monitor the deformation of the reservoir bank slope under the condition of water level decline and track its effect in real time after the early warning treatment measures are implemented, dynamic adjustment can be carried out through an intelligent monitoring data fusion and feedback optimization system. First, the software system needs to integrate the data of InSAR images, GNSS monitoring points, tilt sensors, acceleration sensors, and water level monitoring stations, and use time series analysis and state estimation algorithms (such as Kalman filtering, exponential smoothing, LSTM deep learning model) to continuously track the deformation trend of the slope. The system can automatically calculate the optimal sampling frequency of different monitoring devices according to the correlation between the water level change rate and historical instability events, and apply a dynamic adaptive data sampling strategy (such as an anomaly detection trigger mechanism). When the SEC value continuously approaches the high-risk interval, increase the GNSS sampling frequency and InSAR image update frequency to more accurately capture the dynamic deformation of the slope. At the same time, the system can call a real-time data correlation analysis model to evaluate the time lag and mutual correlation between different data sources, so as to optimize the data fusion strategy and improve the accuracy of real-time monitoring. In addition, the software system can adopt a feedback control mechanism. After the early warning treatment measures are implemented, compare the monitoring data before and after implementation to judge whether there is a trend of risk reduction; if the SEC still remains at a high risk or continues to rise after the early warning measures are implemented, trigger a higher-level monitoring strategy and notify the management department to re-evaluate the effectiveness of the current early warning measures.

[0105] When the system detects that the evaluation result of the instantaneous slope instability remains in a high-risk state, it is necessary to adjust the monitoring strategy and dynamically optimize the early warning mechanism, which can be achieved through dynamic threshold adjustment based on machine learning and statistical optimization. The software system can call adaptive threshold adjustment algorithms (such as quantile regression analysis, Bayesian optimization, clustering analysis) to evaluate the volatility of the SEC calculation results in real time, and combine historical instability cases to automatically adjust the preset threshold of SEC and optimize the risk classification criteria. In addition, the system can apply reinforcement learning algorithms to optimize the early warning rules and improve the adaptability of the early warning model by iteratively analyzing the relationship between the monitoring data and the execution effect of the early warning measures. For example, when the SEC is in a high-risk state multiple times and the current early warning measures fail to effectively reduce the risk, the system can automatically trigger new emergency measures, such as further adjusting the reservoir operation, increasing slope drainage measures, or deploying more sensors to monitor key areas. At the same time, the software can integrate multi-objective optimization algorithms to dynamically allocate monitoring resources according to the power consumption of monitoring devices, data acquisition costs, and early warning accuracy, ensuring that the early warning response speed and accuracy are improved without wasting computing resources. The purpose of this is to ensure that during the long-term water level decline, the slope risk assessment can be dynamically adjusted, avoiding the failure of early warning measures and improving the effectiveness of disaster prevention and mitigation.

[0106] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula that is closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0107] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0108] It should be understood that in various embodiments of the present application, the magnitudes of the serial numbers of the above processes do not imply the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0109] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0110] In several embodiments provided by the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the above-described embodiments are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of devices or units can be in electrical, mechanical or other forms.

[0111] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0112] In addition, the functional units in various embodiments of the present application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0113] As described above, only the specific implementation manners of the present application are described, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for processing the deformation monitoring and early warning of the reservoir bank slope based on InSAR data, characterized in that Specifically, it includes the following steps: In the reservoir bank slope area, obtain the water level change information of the reservoir bank slope in real time through InSAR image data and water level monitoring data, and analyze it after obtaining. Judge whether the water level drop rate exceeds the preset threshold. If the water level drop rate exceeds the preset threshold, it is determined that the current situation is a sharp drop in the reservoir water level within a short time; In the case of determining a sharp drop in the reservoir water level within a short time, obtain the slope deformation characteristic information affected by the sudden drop in water level in real time, and analyze it after obtaining, and generate the sudden drop in water level deformation response coefficient and the instantaneous instability trend index respectively; Construct an instantaneous instability risk assessment model for the generated sudden drop in water level deformation response coefficient and instantaneous instability trend index, and generate the slope instantaneous instability assessment coefficient; Based on the slope instantaneous instability assessment coefficient, conduct an assessment and analysis to evaluate the slope instantaneous instability risk in the case of a sharp drop in the reservoir water level within a short time, and classify it into low risk, medium risk and high risk; According to the assessment results, implement corresponding early warning treatment measures for different risk levels; Continuously monitor the deformation of the reservoir bank slope under the condition of water level drop, and track its effect in real time after the implementation of the early warning treatment measures. If the slope instantaneous instability assessment result shows that there is still a high risk, adjust the monitoring strategy and dynamically optimize the early warning mechanism.

2. The method for monitoring, warning and processing the deformation of the reservoir bank slope based on InSAR data according to claim 1, wherein In the case of determining a sharp drop in the reservoir water level within a short time, obtain the slope deformation characteristic information affected by the sudden drop in water level in real time, and analyze it after obtaining, and generate the sudden drop in water level deformation response coefficient and the instantaneous instability trend index respectively. Specifically, it includes the following steps: In the case of determining a sharp drop in the reservoir water level within a short time, obtain the slope deformation characteristic information affected by the sudden drop in water level in real time, and preprocess it after obtaining; Extract the sudden drop in water level deformation characteristic information and slope instability evolution information from the preprocessed slope deformation characteristic information; Analyze the extracted sudden drop in water level deformation characteristic information and slope instability evolution information, and generate the sudden drop in water level deformation response coefficient and the instantaneous instability trend index respectively.

3. A method for monitoring and warning the deformation of the reservoir bank slope based on InSAR data according to claim 2, characterized in that, The acquisition logic of the sudden drop in water level deformation response coefficient is as follows: Extract the sudden drop in water level deformation characteristic information from the preprocessed slope deformation characteristic information, specifically including the water level drop rate at different time points within a period during the reservoir water level drop process, the surface displacement change rate of the slope in the horizontal distance, and the deformation acceleration of the slope surface, and represent them respectively by functions RW(t), GD(t) and VS(t) according to the time series, where t is the time point, RW(t) represents the water level drop rate at time point t within a period during the reservoir water level drop process, GD(t) represents the surface displacement change rate of the slope in the horizontal distance at time point t within a period during the reservoir water level drop process, VS(t) represents the deformation acceleration of the slope surface at time point t within a period during the reservoir water level drop process, and the defined time period is [t1, t2]; Calculate the sudden drop in water level deformation response coefficient. The specific calculation formula is as follows: In the formula, CRD is the sudden drop in water level deformation response coefficient.

4. A method for monitoring and warning the deformation of the reservoir bank slope based on InSAR data according to claim 3, characterized in that, The acquisition logic of the instantaneous instability trend index is as follows: Extract the slope instability evolution information from the preprocessed slope deformation characteristic information, specifically including the average displacement rate of the deep part of the slope at different time points within a period during the reservoir water level decline, the stress change rate of the slope body, and the acceleration of the deformation rate on the slope surface, and represent them respectively by functions VD(t), SS(t), and AV(t) according to the time series. t is the time point. VD(t) represents the average displacement rate of the deep part of the slope at time point t within a period during the reservoir water level decline. SS(t) represents the stress change rate of the slope body at time point t within a period during the reservoir water level decline. AV(t) represents the acceleration of the deformation rate on the slope surface at time point t within a period during the reservoir water level decline. Define the time period as [t1, t2]. Calculate the instantaneous instability trend index. The specific calculation formula is as follows: In the formula, ITI is the instantaneous instability trend index.

5. A method for monitoring and warning of the deformation of the reservoir bank slope based on InSAR data according to claim 4, characterized in that, Construct an instantaneous instability risk assessment model for the generated rapid drawdown deformation response coefficient CRD and instantaneous instability trend index ITI, and generate a slope instantaneous instability assessment coefficient through weighted summation. The specific calculation formula is as follows: SEC = ω1 * CRD + ω2 * ITI In the formula, SEC is the slope instantaneous instability assessment coefficient. ω1 and ω2 are non-zero weight coefficients of the rapid drawdown deformation response coefficient CRD and the instantaneous instability trend index ITI respectively, and ω1 + ω2 = 1.

6. A method for monitoring and warning the deformation of the reservoir bank slope based on InSAR data according to claim 5, characterized in that, Determine the threshold interval SEC of the preset instantaneous slope instability evaluation coefficient min , SEC max , and compare it with the generated instantaneous slope instability evaluation coefficient SEC after determination. According to the comparison results, evaluate the instantaneous slope instability risk under the condition of a sharp drop in the reservoir water level in a short time, and classify it into low risk, medium risk and high risk. The specific comparison and analysis are as follows: If SEC < SEC min , the risk of instantaneous slope instability under the condition of a sharp drop in reservoir water level within a short time is low risk; If SEC min ≤SEC≤SEC max , the risk of instantaneous slope instability in the case of a sharp drop in reservoir water level within a short period of time is medium risk; If SEC > SEC max , the risk of instantaneous slope instability in the case of a sharp drop in reservoir water level within a short time is a high risk.

7. A method for monitoring and warning the deformation of the reservoir bank slope based on InSAR data according to claim 6, characterized in that, According to the evaluation results, implement corresponding early warning treatment measures for different risk levels. Specifically: For the case where the evaluation result is high risk, the specific early warning treatment measures are: immediately activate the high-level landslide early warning mechanism, and send high-priority early warning information to the management department, emergency rescue units, and surrounding residents; Synchronously adjust the monitoring mode to increase the acquisition frequency of InSAR data; take temporary slope reinforcement measures; at the same time, adjust the reservoir operation strategy to reduce the risk of further inducing instability, and continuously monitor the slope state; For the case where the evaluation result is medium risk, the specific early warning treatment measures are: trigger the conventional slope early warning mechanism, send medium-priority early warning information to the monitoring unit and the reservoir management party, and remind to strengthen monitoring; dynamically adjust the monitoring frequency, increase the update frequency of InSAR images, and increase the monitoring intensity of surface strain sensors; optimize the water level decline rate in combination with the reservoir operation plan; adjust the threshold update strategy of the early warning system; For the case where the evaluation result is low risk, the specific early warning treatment measures are: maintain the current monitoring strategy unchanged, and keep the conventional monitoring frequency of InSAR image update and surface strain monitoring; continuously track the water level decline rate and slope deformation trend to ensure the stability of the evaluation result.

Images