A river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement

Abstract

The present invention discloses a riverbank collapse early warning method based on the analysis of the acceleration trend of soil surface displacement, which relates to the technical field of riverbank collapse early warning. The method includes: arranging GNSS measurement site positions according to the potential risk areas of riverbank beach collapses and the transverse and longitudinal distances of collapse pits in the region; performing real-time traceability filtering and denoising calculations on the continuously online monitored soil surface displacement data based on a Butterworth low-pass filter; performing real-time acceleration trend discrimination calculations on the filtered soil surface displacement data, and using the calculated linear acceleration trend intensity to perform risk classification early warning discrimination on the collapse risk thresholds of different collapse risk levels; carrying out on-site surveys and result verification of riverbank collapse risks according to the results of the risk classification early warning discrimination. The method of the present invention has important engineering application values in aspects such as river safety monitoring, disaster early warning, and emergency management, and can provide scientific support for river regulation, dike safety maintenance, and the construction of smart water conservancy.

Description

Technical Field

[0001] The present invention relates to the technical field of river bank collapse warning, and particularly relates to a river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement. Background Art

[0002] River bank collapse is a common river disaster phenomenon, which seriously affects river flood control safety, levee stability, and the coastal ecological environment. Especially in large river basins, river bank collapse not only threatens flood control dams, port terminals, and coastal infrastructure, but also may cause potential safety hazards to ship navigation. Therefore, timely and accurately monitoring the collapse of river bank beaches and giving early warnings of bank collapse is of great significance for disaster prevention and reduction.

[0003] Traditional bank collapse warning methods mainly include near-shore underwater topographic survey, manual inspection of the bank beach, and UAV inspection. Although these methods can provide certain bank beach deformation conditions, they have limitations in the real-time and continuity of data acquisition and are difficult to meet the warning requirements for sudden bank collapse disasters. In recent years, the rapid development of GNSS (Global Navigation Satellite System) high-frequency online monitoring technology has made it possible to dynamically monitor the soil surface displacement of river bank beaches with millimeter-level accuracy.

[0004] However, due to the complexity of river bank collapse and the fact that high-frequency GNSS measurement data contains a large amount of environmental noise and systematic errors, directly using it for bank collapse warning is very likely to lead to misjudgment or missed judgment. Therefore, how to reasonably deploy GNSS in potential bank collapse areas, how to effectively denoise GNSS high-frequency monitoring data and extract real trend displacement signals, how to identify different acceleration trends under different risk omens before collapse, and thus establish a real-time hierarchical discrimination bank collapse risk model based on soil surface displacement online monitoring data is a technical problem that needs to be solved urgently. Summary of the Invention

[0005] In view of the above problems, the present invention proposes a river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement, aiming to achieve real-time, accurate, and intelligent bank collapse risk assessment and warning.

[0006] The present invention realizes the above object through the following technical solutions:

[0007] A river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement, the method includes:

[0008] According to the potential risk area of river bank beach collapse in the region and the transverse and longitudinal distances of the collapse pit, GNSS measurement site positions are arranged for real-time collection of soil surface displacement data of the site positions;

[0009] Design a low-pass filtering algorithm based on a Butterworth low-pass filter to perform real-time traceability filtering and denoising calculations on the continuously monitored soil surface displacement data;

[0010] According to the linear fitting method, perform real-time acceleration trend discrimination calculations on the filtered soil surface displacement data, and use the calculated linear acceleration trend intensity to perform risk classification early warning discrimination on the bank collapse risk thresholds for different bank collapse risk levels;

[0011] Carry out on-site inspections and result verification of the river bank collapse risk according to the results of the risk classification early warning discrimination, use the results of the on-site inspections to verify the accuracy of the risk classification early warning discrimination, and further optimize the surface displacement threshold of the soil surface displacement.

[0012] As a preferred embodiment of the present invention, the method for arranging GNSS measurement stations includes:

[0013] Divide the general monitoring area and the key monitoring area according to the slope ratio of the bank slope and the height difference between the beach and the trough. The key monitoring area refers to the area where the slope ratio of the bank slope S ≤ 1:3 or the height difference between the beach and the trough H ≥ 20 meters. The area outside the key monitoring area is the general monitoring area;

[0014] In the direction along the shoreline, the layout density of the GNSS measurement stations is determined according to the transverse scale characteristics of the on-site collapse, and the layout spacing L log ≤ 2B, where B represents the continuous transverse distance of the bank collapse or the transverse width of the collapse pit;

[0015] In the direction perpendicular to the shoreline, the layout density of the GNSS measurement stations is determined according to the longitudinal distance characteristics of the on-site collapse, and the longitudinal layout spacing L lat ≤ 2D, where D represents the average recession distance of the bank collapse or the longitudinal scale of the collapse pit;

[0016] The GNSS measurement stations are arranged in the beach area between the river dike and the water edge line, and the layout elevation is lower than the dike elevation and is between the dike elevation and the average high water level of the river.

[0017] As a preferred embodiment of the present invention, the expression of the soil surface displacement data is:

[0018] ;

[0019] In the formula, is the time series data of the high-frequency soil surface displacement monitored by the GNSS measurement station; represents the low-frequency trend soil displacement part; represents the random noise part;

[0020] Use the Butterworth low-pass filter to design a low-pass filtering algorithm to perform forward traceability filtering and denoising calculations on the random noise part for.

[0021] As a preferred embodiment of the present invention, the low-pass filtering algorithm is characterized by two parameters, namely the order of the filter N and the cut-off frequency at -3 dB ;

[0022] For the soil surface displacement data monitored by the GNSS measurement station, the normalized amplitude-frequency response function of the Butterworth filter is expressed as:

[0023] ;

[0024] In the formula, is the signal frequency, is the cut-off frequency, N is the order of the filter.

[0025] As a preferred embodiment of the present invention, the order of the filter N is determined according to the following formula:

[0026] ;

[0027] In the formula, is the passband attenuation; is the stopband attenuation; is the sampling frequency;

[0028] The cut-off frequency is determined by the following method:

[0029] For the time series data of the soil surface displacement perform spectral analysis using the fast Fourier transform to calculate the corresponding spectrum :

[0030] ;

[0031] In the formula, M is the number of points of the fast Fourier transform, is the imaginary unit; is the frequency index representing the frequency component in the spectrum, is the time index representing the sampling points in the time series;

[0032] Calculate the spectrum amplitude :

[0033] ;

[0034] In the formula, and are the real part and the imaginary part after the fast Fourier transform respectively;

[0035] By analyzing the spectral amplitude, the cumulative contribution rate of the frequency energy is statistically calculated , and the formula is:

[0036] ;

[0037] In the formula, represents the index corresponding to the candidate cut-off frequency; M / 2 represents the maximum index in the positive frequency range;

[0038] Select the minimum frequency that satisfies the cumulative contribution rate as the cut-off frequency .

[0039] As a preferred solution of the present invention, the real-time acceleration trend discrimination calculation is performed on the filtered soil surface displacement, and the risk classification early warning discrimination is performed on the surface displacement thresholds of different bank collapse risk levels by using the calculated linear acceleration trend intensity, specifically including:

[0040] Perform real-time acceleration trend discrimination calculation on the soil surface displacement after denoising filtering according to the least square method, and calculate the linear increasing rate of the trend displacement, which represents the moving speed of the monitored displacement on the soil surface during collapse ;

[0041] When calculating the linear increasing rate of the trend displacement, select the data of the continuous 5 hours before the displacement judgment time point as the calculation period, denoted as , and the sampling frequency of the GNSS measurement station is ;

[0042] For time, in order to calculate the moving speed of the soil surface displacement and discriminate the bank collapse risk, the number of the previous data points in the continuous 5 hours pushed forward from time is used in the calculation as:

[0043] ;

[0044] Carry out the least square method calculation for the previous sampling data:

[0045] ;

[0046] In the formula, is the corresponding time of the sampling point in the calculation domain pushed forward from the current calculation time point ; is the soil surface displacement corresponding to the sampling point in the calculation domain pushed forward from the current calculation time point .

[0047] As a preferred embodiment of the present invention, the bank collapse risk thresholds for different bank collapse risk levels are set according to different risk situations of riverbank collapse, specifically divided into:

[0048] Low-risk state, i.e., safety zone: The trend deformation is within the set acceptable range, approaching the static stability state. At this time, the bank collapse risk threshold is denoted as , and the corresponding trend deformation rate range is ;

[0049] Medium-risk state, i.e., warning zone: The trend deformation has a cumulative trend, cracks occur inside the soil mass, and the cracks continue to extend. At this time, the bank collapse risk threshold is denoted as , and the corresponding trend deformation rate range is , corresponding to issuing a medium-risk warning for riverbank collapse;

[0050] High-risk state, i.e., critical zone: After the surface displacement of the soil mass lasts for a preset time period in the warning zone, The trend deformation has an accelerating development trend. At this time, through-cracks appear inside the soil mass, and local soil masses reach the critical sliding state, facing the situation of instability and collapse. Moreover, after the collapse, the entire collapsed body will quickly lose its bearing capacity. The corresponding trend deformation rate range is , corresponding to issuing a high-risk warning for riverbank collapse.

[0051] As a preferred embodiment of the present invention, on the basis of the results of risk classification early warning discrimination, on-site investigation and result verification of riverbank collapse risk are carried out, and the accuracy of risk classification early warning discrimination is verified by using the results of on-site investigation, and the bank collapse risk threshold of soil surface displacement is further optimized, specifically including:

[0052] Early warning response investigation: When the trend deformation of the soil surface displacement is in the medium-risk state or high-risk state and an early warning is issued, organize an on-site investigation to observe the cracks on the bank slope, including the width, depth, and trend of the cracks, check for signs of local collapse, and evaluate the threat of bank collapse;

[0053] On-site data comparison: Compare the on-site measured data, including the landslide location and the crack expansion situation, with the predicted early warning state in the real-time hierarchical discrimination result of the acceleration trend of the soil surface displacement;

[0054] If the on-site bank collapse situation is stronger than the early warning state, it is necessary to lower the critical bank collapse risk threshold;

[0055] If the on-site bank collapse situation is lower than the early warning state, it is necessary to raise the critical bank collapse risk threshold.

[0056] By deploying GNSS measurement stations in the potential risk areas of riverbank collapses, it is possible to collect real-time soil surface displacement data, providing real-time and accurate data support for the early identification and warning of bank collapse risks. The Butterworth low-pass filter is used to denoise the soil surface displacement data, significantly reducing environmental interference and measurement errors, improving the accuracy of the data, and ensuring the reliability of subsequent trend analysis. By analyzing the filtered displacement data through linear fitting methods, the acceleration trend intensity is calculated in real time, and risk-based grading warnings for bank collapses are issued based on this trend, effectively improving the accuracy and response speed of the warnings. Field inspections are carried out according to the results of the risk-based grading warnings, and the accuracy of the warnings is verified by combining the measured data, further optimizing the risk thresholds to make the warning system more accurate and operable. The results of the field inspections and verifications provide a basis for optimizing the warning thresholds, ensuring that the warning system can be dynamically adjusted according to the actual situation, thereby improving the adaptability and long-term effectiveness of the system. Through the combination of technical means such as GNSS real-time monitoring, low-pass filter denoising, acceleration trend analysis, risk-based grading warnings, and field verification, the present invention forms an efficient, intelligent, and accurate method for warning riverbank collapses, capable of capturing real-time changes in soil displacement and conducting scientific risk assessments, providing strong support for the early warning and prevention of bank collapse disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. Among them:

[0058] Figure 1 is the flowchart of the method of the present invention;

[0059] Figure 2 is the schematic diagram of the on-site layout of GNSS measurement stations in the embodiments of the present invention;

[0060] Figure 3 is the comparison diagram of the original soil surface displacement data sequence and the displacement data sequence after Butterworth low-pass filtering in the embodiments of the present invention;

[0061] Figure 4 is the schematic diagram of the risk-based grading warning discrimination for the acceleration change trend of soil surface displacement in the embodiments of the present invention;

[0062] Figure 5 is the schematic diagram of the on-site photo of the low-risk state of bank collapse in the embodiments of the present invention;

[0063] Figure 6 is the schematic diagram of the on-site photo of the high-risk state of bank collapse in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions of the embodiments of the present invention in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention fall within the scope of protection of the present invention.

[0065] As Figure 1 shown, it is an embodiment of the present invention. This embodiment provides a riverbank collapse warning method based on the analysis of the acceleration trend of soil surface displacement, including the following steps:

[0066] S1: Horizontally and vertically arrange GNSS (Global Navigation Satellite System) measurement stations

[0067] According to the potential risk area of riverbank beach collapse in the region and the transverse and longitudinal distances of the collapse pits, reasonably arrange the positions of GNSS measurement stations for real-time collection of soil surface displacement data of the points.

[0068] Specifically, the method for arranging the positions of GNSS measurement stations includes:

[0069] Divide the general monitoring area and the key monitoring area according to the slope ratio of the riverbank slope and the elevation difference between the beach and the trough. The spacing of GNSS measurement stations in the key monitoring area is denser. The key monitoring area refers to the area where the slope ratio S of the riverbank slope ≤ 1:3 or the elevation difference H between the beach and the trough ≥ 20 meters. The area outside the key monitoring area is the general monitoring area;

[0070] In the direction along the shoreline, the layout density of GNSS measurement stations is determined according to the transverse scale characteristics of the on-site collapse. The layout spacing L log ≤ 2B, where B represents the transverse distance of continuous riverbank collapse or the transverse width of the collapse pit; for example, for a large-scale collapse pit in a certain section, the transverse width can reach about 500m, so the layout spacing can be taken as 1km;

[0071] In the direction perpendicular to the shoreline, the layout density of GNSS measurement stations is determined according to the longitudinal distance characteristics of the on-site collapse. The longitudinal layout spacing L lat ≤ 2D, where D represents the average recession distance of riverbank collapse or the longitudinal scale of the collapse pit; for example, for a large-scale collapse pit in a certain section, the average recession depth can reach 60m, so the layout spacing can be taken as 120m;

[0072] The GNSS measurement stations are arranged in the beach area between the river dike and the water edge line. This area has been eroded and collapsed under the condition of water flow scouring, and at the same time, it can take into account the requirements of convenient construction and long-term maintenance. Generally speaking, the elevation of the equipment layout is lower than the elevation of the dike and is between the elevation of the dike and the average high water level of the river (the average high tide level is taken for tidal reaches).

[0073] As shown Figure 2 in the figure, an on-line measurement station for GNSS soil surface displacement measurement is arranged on the water-facing beach of the water-emerging sandbank in a certain section of the river. The average high tide level here is about +2m, and the embankment elevation is about +5m. Therefore, the elevation of the measurement station arrangement area is +3m.

[0074] S2: High-frequency on-line measurement displacement data spectrum analysis and denoising calculation

[0075] Based on the Butterworth low-pass filter, a low-pass filtering algorithm is designed to perform real-time traceability filtering and denoising calculation on the continuously on-line monitored soil surface displacement data.

[0076] In one embodiment, the expression of the soil surface displacement data is:

[0077] ;

[0078] where is the time series data of the high-frequency soil surface displacement monitored by the GNSS measurement station, taking the synthetic displacement in the three-dimensional direction, and the sampling frequency is , generally 1 data per 1 - 3 minutes; represents the low-frequency trend soil displacement part; represents the random noise part, which is mainly caused by factors such as measurement error, environmental interference, and instrument stability, including receiver white noise, tropospheric and ionospheric delay noise, and noises such as vehicle passing, wind action, and temperature change in the observation environment;

[0079] Use the Butterworth low-pass filter to design a low-pass filtering algorithm to perform forward traceability filtering and denoising calculation on the random noise part , where the low-pass filtering algorithm needs to be characterized by two parameters, namely the order of the filter N and the cut-off frequency at -3dB ;

[0080] For the soil surface displacement data monitored by the GNSS measurement station, the normalized amplitude-frequency response function of the Butterworth filter is expressed as:

[0081] ;

[0082] where is the signal frequency, is the cut-off frequency, N is the order of the filter;

[0083] When , The signal has almost no attenuation;

[0084] When then the signal is effectively attenuated;

[0085] The order of the filter N is larger, and the transition band (the region from 1 to 0) of the filter is steeper.

[0086] According to the general characteristics of the surface displacement change of the soil body during bank collapse, the order of the Butterworth low-pass filter N is selected to be more suitable for the medium order, achieving a balance between the smoothing effect and the signal fidelity;

[0087] Furthermore, the order of the filter N is determined according to the following formula:

[0088] ;

[0089] wherein, is the passband attenuation, with the unit of dB, and the default value is ; is the stopband attenuation, with the unit of dB. For the surface displacement data of the soil body, the default value is ; is the sampling frequency; the sampling frequency of this data sequence is 1 data per 1 minute, and after conversion, the frequency is 1 / 60 Hz.

[0090] The cut-off frequency is selected based on the spectral characteristics of the actual GNSS soil surface displacement signal, and the FFT (Fast Fourier Transform) spectral analysis method combined with the energy contribution method can be used to determine it;

[0091] Specifically, the cut-off frequency is determined according to the following method:

[0092] Perform spectral analysis on the time series data of the soil surface displacement using the Fast Fourier Transform to calculate the corresponding spectrum :

[0093] ;

[0094] wherein, M is the number of points of the Fast Fourier Transform, is the imaginary unit; is the frequency index representing the frequency component in the spectrum, is the time index representing the sampling points in the time series;

[0095] Calculate the spectrum amplitude :

[0096] ;

[0097] In the formula, and are the real and imaginary parts after fast Fourier transform respectively;

[0098] By analyzing the spectrum amplitude, the time series data of soil surface displacement is determined In which frequency ranges is the main energy of the signal concentrated? Specifically, the cumulative contribution rate of statistical frequency energy , the formula is:

[0099] ;

[0100] In the formula, Indicates the index corresponding to the candidate cutoff frequency; M / 2 indicates the maximum index of the positive frequency range;

[0101] Choose to meet the cumulative contribution rate The minimum frequency is used as the cut-off frequency In this embodiment, the cut-off frequency can be taken as 0.03 Hz to 0.05 Hz according to calculation.

[0102] like Figure 3 The figure shows the comparison between the original high-frequency data sequence of the GNSS measurement station during operation and the displacement data sequence after Butterworth low-pass filtering. The original high-frequency data in the upper part of the figure fluctuates frequently and with large amplitudes, indicating that the data is affected by more high-frequency noise, which may be caused by the external environment (such as wind, temperature changes, etc.) or measurement errors. The fluctuation pattern of the original data is relatively complex, and it is difficult to extract a clear trend displacement from it.

[0103] The data below, after Butterworth filtering, has significantly reduced volatility, showing a smooth and gradually changing curve. After filtering, most of the high-frequency noise is removed, making the data closer to the actual trend displacement. The filtered data shows a more obvious and continuous change trend (such as a gentle curve of rising or falling), which shows that the long-term trend displacement is effectively extracted through the low-pass filter, and irrelevant high-frequency fluctuations are removed, which can help better analyze the long-term changes in soil displacement.

[0104] S3: Risk classification and early warning of soil surface displacement acceleration trend

[0105] According to the linear fitting method, real-time acceleration trend discrimination calculation is carried out on the filtered soil surface displacement data, and the risk classification early warning discrimination of the bank collapse risk threshold for different bank collapse risk levels is carried out by using the calculated linear acceleration trend intensity.

[0106] In one embodiment, the implementation method of step S3 includes:

[0107] Perform real-time acceleration trend discrimination calculation on the soil surface displacement after denoising filtering according to the least square method, and calculate the linear increasing rate of the trend displacement, which represents the moving speed of the monitored soil surface displacement during collapse. ;

[0108] According to the general characteristics of river bank collapse, within 3 to 5 hours before the bank collapse, the soil displacement shows a significant trend of cumulative increase. Therefore, when calculating the linear increasing rate of the trend displacement, select the data of the continuous 5 hours before the displacement judgment time point as the calculation period, denoted as , and the sampling frequency of the GNSS measurement station is (1 sampling data per minute);

[0109] For moment, in order to calculate the moving speed of the soil surface displacement and discriminate the bank collapse risk, the number of previous data points within the continuous 5 hours pushed forward from moment is used for calculation, denoted as :

[0110] ;

[0111] Carry out least square method calculation for the previous data:

[0112] ;

[0113] In the formula, is the corresponding time of the sampling point within the calculation domain pushed forward from the current calculation time point ; is the soil surface displacement corresponding to the sampling point within the calculation domain pushed forward from the current calculation time point .

[0114] As Figure 4 shown, there are 40,800 data points in the example calculation process, and the data duration is 20 days.

[0115] The bank collapse risk thresholds for different bank collapse risk levels are set according to different risk situations of river bank collapse, and are specifically divided into:

[0116] Low risk state (safe area): The trend deformation is within the set acceptable range and is close to the static and stable state. At this time, the bank collapse risk threshold is denoted as , the corresponding trend deformation rate range is ; in this calculation takes the value of 0;

[0117] Medium risk state (warning area): The trend deformation has a cumulative trend. Generally, cracks occur inside the soil mass and the cracks continue to extend. At this time, the bank collapse risk threshold is denoted as , the corresponding trend deformation rate range is , corresponding to issuing a medium risk warning for river bank collapse; in this calculation takes the value of 0.0005;

[0118] High risk state (critical area): After the soil surface displacement continues in the warning area for a preset time period, the trend deformation has an obvious accelerating development trend. At this time, through cracks appear inside the soil mass, and local soil reaches the critical sliding state, facing the situation of instability and collapse. And after the collapse, the entire collapsed body will quickly lose its bearing capacity. The corresponding trend deformation rate range is , corresponding to issuing a high risk warning for river bank collapse.

[0119] Among them, the bank collapse risk thresholds of different bank collapse risk levels 、 The value-taking process is determined by comprehensively considering the bank collapse characteristics, geological conditions, protection conditions, historical bank collapse characteristics and other conditions in different regions.

[0120] Such as Figure 4 shown, it is the risk level warning discrimination result based on the accelerating change trend of the soil surface displacement of the example measurement data during the operation process. It can be seen from the figure that from April 24th to May 5th, the fluctuating change of the soil surface displacement data is generally stable, and the discrimination result is in the low risk state of bank collapse; from May 5th to May 12th, the soil surface displacement data increases, and the discrimination result is in the medium risk state of bank collapse; from May 12th to May 15th, the soil surface displacement data accelerates to increase, and the discrimination result is in the high risk state of bank collapse.

[0121] S4: On-site inspection and verification optimization of the risk classification threshold for river bank collapse after warning

[0122] According to the results of the risk classification warning discrimination, conduct on-site inspection and result verification of the river bank collapse risk, use the results of the on-site inspection to verify the accuracy of the risk classification warning discrimination, and further optimize the bank collapse risk threshold of the soil surface displacement.

[0123] Specifically, step S4 includes:

[0124] Early warning response on-site inspection: When early warnings are issued due to the trend of medium and high risks in the surface displacement of the soil mass, organize on-site inspections to observe the bank slope cracks (width, depth, and direction), check for signs of local collapse (collapse, sliding soil blocks), and assess the threat of bank collapse.

[0125] On-site data comparison: Compare the on-site measured data, including the landslide location, crack expansion, and the predicted early warning status in the real-time hierarchical discrimination results of the acceleration trend of the soil mass surface displacement. If the on-site bank collapse situation is stronger than the early warning status, it indicates that the early warning model is too conservative and the critical bank collapse risk threshold needs to be lowered; if the on-site bank collapse threat is lower than the early warning status, the critical bank collapse risk threshold needs to be further increased.

[0126] Verify the accuracy of the discrimination results through the results of multiple on-site inspections, further optimize the bank collapse risk threshold of the soil mass surface displacement, enhance the scientific nature of the use of the early warning method, and reduce false alarms and missed alarms.

[0127] On April 30th and May 15th, on-site inspections were carried out to check the collapse situation of the river channel where the GNSS measurement station is located. The on-site photos are shown in Figure 5 and Figure 6 . On April 30th, there was no obvious erosion at the river channel front, the beach was relatively stable, and there were no cracks in the on-site soil mass, being in a low-risk state of bank collapse; on May 15th, a penetrating large crack appeared at the beach front, and the front was severely eroded, truly facing a high-risk state of imminent bank collapse. This early warning method was verified to work well in the on-site environment.

[0128] In summary, through the layout of GNSS measurement stations and the combination of real-time online monitoring data of the soil mass surface displacement, the present invention can accurately capture the displacement changes of river channel bank collapse in real time, ensure timely early warnings before the occurrence of bank collapse, effectively remove high-frequency noise using a Butterworth low-pass filter, significantly improve the accuracy of displacement data, and thus enhance the accuracy of early warnings;

[0129] By denoising the soil mass surface displacement data and performing acceleration trend discrimination based on the linear fitting method, the acceleration trend of the soil mass surface can be analyzed, and the potential risks of bank collapse can be accurately identified; through acceleration trend analysis, the linear growth rate of soil mass displacement can be revealed, and risk grading can be carried out according to different speed ranges, effectively enhancing the response speed and accuracy of bank collapse risk early warnings;

[0130] By setting different risk level thresholds (low risk, medium risk, high risk), detailed grading is carried out for different soil mass displacement trends, enabling the early warning system to issue corresponding early warning information according to different risk states; combined with on-site inspections and result verification, the bank collapse risk threshold is further optimized. This verification process can adjust and improve the reliability of the early warning model in real time, and enhance the effectiveness and pertinence of early warnings;

[0131] The hierarchical early warning scheme and on-site verification mechanism proposed by the present invention can not only achieve intelligent decision-making, but also dynamically adjust according to the on-site environment and actual situation, enhancing the scientificity and practicality of bank collapse prevention and control; through continuous optimization and real-time data feedback, it can automatically adjust the bank collapse risk threshold, further improving the accuracy and timeliness of bank collapse early warning.

[0132] The present invention provides an efficient, intelligent, and accurate river bank collapse early warning method through precise displacement monitoring, effective noise filtering, acceleration trend discrimination, and multi-level risk hierarchical early warning, greatly improving the early warning ability and prevention and control effect of river bank collapse disasters.

[0133] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, it can be implemented in whole or in part in the form of a computer program product, which includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to 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.

[0134] In addition, in each embodiment of the present application, the functional units can be integrated into a processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. When the above integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a magnetic disk, or an optical disc, etc.

[0135] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various changes or substitutions, which should all 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 warning of river bank collapse based on the analysis of the acceleration trend of soil surface displacement, characterized in that The method includes: According to the potential risk areas of riverbank collapses and the lateral and longitudinal distances of collapse pits in the region, GNSS measurement station positions are arranged. Areas with a slope ratio S of the bank slope ≤ 1:3 or a beach-trough height difference H ≥ 20 meters are defined as key monitoring areas. The layout spacing L of GNSS measurement stations along the coastline direction log ≤ 2B, and the longitudinal layout spacing L lat ≤ 2D, which is used to collect the soil surface displacement data of the point positions in real time; where B represents the lateral distance of continuous bank collapses or the lateral width of collapse pits, and D represents the average recession distance of bank collapses or the longitudinal scale of collapse pits; Design a low-pass filtering algorithm based on Butterworth low-pass filter to perform real-time traceability filtering and denoising calculations on continuously monitored soil surface displacement data; the low-pass filtering algorithm needs to be characterized by two parameters, namely the order of the filter N and the cut-off frequency at -3dB , and the cut-off frequency is determined by the following method: For the time series data of soil surface displacement Perform spectral analysis using the fast Fourier transform and calculate the corresponding spectrum : ； In the formula, M is the number of points of the fast Fourier transform, is the imaginary unit; is the frequency index representing the frequency component in the spectrum, is the time index representing the sampling points in the time series; Calculate the spectral amplitude : ； In the formula, and are the real part and the imaginary part after the fast Fourier transform, respectively; By analyzing the spectral amplitude, the cumulative contribution rate of the frequency energy is statistically calculated , and the formula is as follows: ； wherein, represents the index corresponding to the candidate cut-off frequency; M / 2 represents the maximum index of the positive frequency range; Select the minimum frequency that satisfies the cumulative contribution rate as the cut-off frequency ; Performing real-time acceleration trend discrimination calculation on the filtered soil surface displacement data according to the linear fitting method, and using the calculated linear acceleration trend intensity to conduct risk classification early warning discrimination on the bank collapse risk thresholds of different bank collapse risk levels; Carrying out on-site investigation and result verification of the river bank collapse risk according to the results of the risk classification early warning discrimination, using the results of the on-site investigation to verify the accuracy of the risk classification early warning discrimination, and further optimizing the surface displacement threshold of the soil surface displacement, specifically including: Early warning response on-site investigation: When the soil surface displacement is in a trending deformation in a medium-risk state or a high-risk state and an early warning is issued, organize an on-site investigation to observe the bank slope cracks, including the width, depth, and trend of the cracks, check for signs of local collapse, and evaluate the bank collapse threat situation; On-site data comparison: Compare the on-site measured data, including the landslide location and the crack expansion situation, with the early warning state predicted in the real-time classification discrimination result of the acceleration trend of the soil surface displacement; If the on-site bank collapse situation is stronger than the early warning state, it is necessary to lower the critical bank collapse risk threshold; If the on-site bank collapse situation is lower than the early warning state, it is necessary to raise the critical bank collapse risk threshold.

2. The river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement as described in claim 1, characterized in that, The GNSS measurement station is arranged in the beach area between the river bank dike and the water edge line, the arranged elevation is lower than the dike elevation, and it is located between the dike elevation and the average high water level of the river.

3. The river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement according to claim 1, wherein The expression of the soil surface displacement data is: ； wherein, is the time series data of the high-frequency soil surface displacement monitored by the GNSS measurement station; represents the low-frequency trend part of the soil displacement; represents the random noise part; Design a low-pass filtering algorithm using a Butterworth low-pass filter to perform forward traceability filtering and denoising calculations on the random noise part for the forward traceability filtering and denoising calculation of the random noise part 4. The river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement according to claim 3, characterized in that, For the soil surface displacement data monitored by the GNSS measurement station, the normalized amplitude-frequency response function of the Butterworth filter is expressed as: ； In the formula, is the signal frequency, is the cut-off frequency, N is the order of the filter.

5. A river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement according to claim 4, characterized in that The order of the filter N is determined by the following formula: ； Wherein, is the passband attenuation; is the stopband attenuation; is the sampling frequency.

6. The river bank collapse warning method based on the analysis of the acceleration trend of soil surface displacement according to claim 1, wherein, Performing real-time acceleration trend discrimination calculation on the filtered soil surface displacement, and using the calculated linear acceleration trend intensity to conduct risk classification early warning discrimination on the surface displacement thresholds of different bank collapse risk levels, specifically including: The real-time acceleration trend discrimination calculation is carried out on the soil surface displacement after denoising filtering according to the least square method, and the linear increase rate of the trend displacement is calculated, which represents the moving speed of the monitored displacement on the soil surface during the collapse. ; When calculating the linear increase rate of the trend displacement, select the data of the continuous 5 hours before the displacement judgment time point as the calculation period, denoted as , and the sampling frequency of the GNSS measurement station is ; For the moment, in order to calculate the moving speed of the soil surface displacement and determine the risk of bank collapse, the number of previous data points within the consecutive 5 hours prior to the moment is used in the calculation as follows: ； Performing least squares calculation on the previous sampling data; ； In the formula, is the corresponding time of the sampling point within the forward calculation domain at the current calculation time point; is the corresponding time of the sampling point within the forward calculation domain at the current calculation time point; is the displacement of the soil surface corresponding to the sampling point within the forward calculation domain at the current calculation time point; is the displacement of the soil surface corresponding to the sampling point within the forward calculation domain at the current calculation time point.

7. The river bank collapse warning method based on the accelerated trend analysis of soil surface displacement according to claim 6, wherein The bank collapse risk thresholds of different bank collapse risk levels are set according to different risk situations of river bank collapse, specifically divided into: Low-risk state, i.e., safety zone: The trend deformation is within the set acceptable range and close to the static stability state. At this time, the bank collapse risk threshold is denoted as , and the corresponding range of trend deformation rate is ; Medium-risk state, i.e., warning area: The trend deformation has a cumulative trend, cracks occur inside the soil mass, and the cracks continue to extend. At this time, the bank collapse risk threshold is recorded as , corresponding to the trend deformation rate range of , corresponding to the issuance of a medium-risk warning for river bank collapse; High-risk state, i.e., critical zone: After the soil surface displacement has persisted in the warning zone for a preset time period, the trend deformation shows an accelerating development trend. At this time, penetrating cracks appear inside the soil mass, and local soil reaches the critical sliding state, facing the situation of instability and collapse. Moreover, the entire collapsed body will quickly lose its bearing capacity after the collapse. The corresponding range of the trend deformation rate is , and a high-risk warning for riverbank collapse is issued accordingly.

Images