High-position landslide stability analysis method based on micro-seismic signal identification

Abstract

The invention relates to the technical field of high-position landslide stability analysis, in particular to a high-position landslide stability analysis method based on micro-seismic signal identification, which comprises the following steps: acquiring a slope rock sample to carry out a destruction experiment, acquiring a test micro-seismic signal by adopting an acoustic emission sensor, constructing a test feature database, and acquiring a fracture dominant frequency feature vector; hydrological characteristic parameters are collected, hydrological characteristic characterization values are calculated, hydrological influence levels of the target area are divided, and the frequency for collecting microseismic signals is determined; actual micro-seismic signals are collected to construct a field feature database, and actual dominant frequency feature vectors are obtained; calculating the similarity between the actual dominant frequency feature vector and the corresponding broken dominant frequency feature vector; judging the high-position landslide stability grade based on the similarity; and based on the high-position landslide stability grade, whether emergency evacuation is triggered or not is determined, and whether drainage equipment and grouting equipment are started or not is determined. According to the invention, the early warning accuracy of the high-position landslide is obviously improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of high landslide stability analysis, and particularly relates to a high landslide stability analysis method based on microseismic signal recognition. BACKGROUND

[0002] With the development of geological disaster monitoring technology, people pay more and more attention to the early warning of high landslides. The existing method collects the microseismic signals generated by the fracture of the slope rock mass, analyzes the signal characteristics to judge the stability of the landslide, and also has a technology that combines multi-source monitoring data such as space, sky, ground and underground, uses statistical methods and machine learning models to comprehensively evaluate landslide risk, and tries to consider the influence of hydrological factors such as rainfall and groundwater level on landslides, providing more comprehensive data support for stability analysis.

[0003] However, the existing technology does not fully consider the influence of different hydrological conditions on the fracture characteristics of rock mass when analyzing microseismic signals. For example, the frequency characteristics of microseismic signals generated during tensile or shear failure of rock mass under different water contents will change, and the traditional method does not establish an effective correction model for such changes. At the same time, when determining the microseismic signal collection frequency and early warning parameters, there is a lack of dynamic adaptation to the hydrological period, which leads to the fact that the monitoring data cannot accurately reflect the actual landslide risk, the accuracy of stability analysis is limited, and it is difficult to achieve reliable early warning under complex hydrological conditions.

[0004] Chinese patent publication No. CN119575460A discloses a slope stability evaluation method based on microseismic signal double-main-frequency mechanism characteristics. The present application relates to material analysis and application, and discloses a slope stability evaluation method based on microseismic signal double-main-frequency mechanism characteristics. The method is based on the objective principle that the main frequency characteristics of rock acoustic emission and the main frequency characteristics of microseismic signals are consistent and related, and the double-main-frequency feature phenomenon exists in acoustic emission test and microseismic monitoring. The main frequency characteristics of the rock sample obtained by acoustic emission test are used to process the microseismic signal waveform data obtained by microseismic monitoring, so that the main frequency characteristics evolution law of the slope rock failure can be obtained through the microseismic signal, and the stress condition of the slope rock can be more accurately judged based on the main frequency characteristics evolution law, and then the stability of the slope is evaluated, so that the accuracy of the evaluation result is significantly improved. The evaluation method combines acoustic emission test and microseismic monitoring, has the advantages of high evaluation accuracy and good universality, and is suitable for large-scale application in the monitoring and prevention of landslides and other geological disasters.

[0005] The Chinese patent publication No. CN115691058A discloses a holographic three-dimensional networking landslide intelligent early warning method based on multiple monitoring occasions. The present application relates to the technical field of early warning methods for geological disasters, in particular to a holographic three-dimensional networking landslide intelligent early warning method based on multiple monitoring occasions. The present application provides a holographic three-dimensional networking landslide intelligent early warning method based on multiple monitoring occasions, which carries out landslide holographic and omnidirectional monitoring in air, space, ground and underground multiple monitoring fields, and constructs a landslide comprehensive early warning criterion coupling multiple characteristics such as surface deformation displacement, rate and acceleration, deep deformation displacement, rate and acceleration, groundwater level, microseismic intensity, rainfall and soil moisture content, effectively overcoming the limitations of large error, low precision and high false alarm of the current single criterion or double criterion, and significantly improving the early warning precision.

[0006] Therefore, the prior art has the following problems: The dynamic changes of the microseismic signal main frequency characteristics under different rainfall conditions are not considered, and the dynamic adjustment of the microseismic signal collection frequency based on the hydrological characteristic parameters is not considered, resulting in low landslide risk identification rate and low early warning accuracy in high-risk hydrological scenarios. SUMMARY

[0007] Therefore, the present application provides a high-position landslide stability analysis method based on microseismic signal identification to overcome the problems that the dynamic changes of the microseismic signal main frequency characteristics under different moisture content conditions are not considered, and the dynamic adjustment of the microseismic signal collection frequency based on the hydrological characteristic parameters is not considered, resulting in low landslide risk identification rate and low early warning accuracy in high-risk hydrological scenarios.

[0008] To achieve the above-mentioned purpose, the present application provides a high-position landslide stability analysis method based on microseismic signal identification, comprising: Collecting the slope rock samples of the target area for failure verification experiments under different moisture content conditions, and collecting the test microseismic signals under each moisture content using acoustic emission sensors; Based on the test microseismic signals under each moisture content, a test feature database is constructed, and a failure main frequency characteristic vector under each moisture content is obtained; Collecting the hydrological characteristic parameters of the target area in the historical period, and calculating the hydrological characteristic representation value based on the hydrological characteristic parameters of the target area in the historical period; Dividing the hydrological influence level of the target area based on the hydrological characteristic representation value, and determining the frequency of collecting microseismic signals based on the hydrological influence level; Collecting the actual microseismic signals of the target area based on the frequency of collecting microseismic signals, constructing a field feature database based on the actual microseismic signals and soil moisture content, and obtaining an actual main frequency characteristic vector; Calculate the similarity between the actual dominant frequency feature vector of the target area and the corresponding fracture dominant frequency feature vector under the corresponding water content, and determine the high-order landslide stability level based on the similarity; Determine whether to trigger emergency evacuation, and determine whether to start the drainage equipment and the grouting equipment based on the high-order landslide stability level; After completing the drainage and grouting adjustment, collect the actual microseismic signal of the adjusted target area, obtain the corresponding actual dominant frequency feature vector, and determine the corresponding high-order landslide stability level based on the similarity; The hydrological characteristic parameters include soil water content, groundwater level change amplitude, and rainfall.

[0009] Further, the process of obtaining the fracture dominant frequency feature vector under each water content based on the test microseismic signal under each water content includes, Generate an energy distribution visualization image in a “time-frequency” two-dimensional coordinate system based on the obtained test microseismic signal under each water content; Obtain the fracture dominant frequency feature vector under each water content based on the energy distribution visualization image; The fracture dominant frequency feature vector includes an energy peak frequency, a ratio of a dominant frequency energy to a total energy, and a frequency band width.

[0010] Further, the process of calculating the hydrological characteristic representation value based on the hydrological characteristic parameters of the target area in the historical period includes, Obtain the soil water content, the groundwater level change amplitude, and the rainfall of the target area in the historical period; Calculate the ratio of the soil water content in a single period to a predetermined soil water content threshold to determine a first hydrological characteristic parameter; Calculate the ratio of the groundwater level change amplitude in a single period to a predetermined groundwater level change amplitude threshold to determine a second hydrological characteristic parameter; Calculate the ratio of the rainfall in a single period to a predetermined rainfall threshold to determine a third hydrological characteristic parameter; Determine the sum of the first hydrological characteristic parameter, the second hydrological characteristic parameter, and the third hydrological characteristic parameter as the hydrological characteristic representation value.

[0011] Further, the process of dividing the hydrological influence level of the target area based on the hydrological characteristic representation value includes, If the hydrological characteristic representation value of the target area is greater than a predetermined hydrological characteristic representation value threshold, it is determined that the hydrological influence level of the target area is a first hydrological influence level; If the hydrological characteristic representation value of the target area is less than or equal to a predetermined hydrological characteristic representation threshold, it is determined that the hydrological influence level of the target area is a second hydrological influence level.

[0012] Further, the process of determining the frequency of collecting microseismic signals based on the hydrological influence level of the target area comprises, If the hydrological influence level of the target area is the first hydrological influence level, the frequency of collecting microseismic signals is determined as a higher frequency; If the hydrological influence level of the target area is the second hydrological influence level, the frequency of collecting microseismic signals is determined as a lower frequency.

[0013] Further, the process of constructing a field feature database based on the actual microseismic signals of the target area and the soil moisture content, and obtaining an actual main frequency feature vector comprises, generating an energy distribution visualization image in a "time-frequency" two-dimensional coordinate system based on the actual microseismic signals of the target area and the soil moisture content obtained; obtaining an actual main frequency feature vector based on the energy distribution visualization image; The actual main frequency feature vector includes an energy peak frequency, a ratio of main frequency energy to total energy, and a frequency band width.

[0014] Further, the process of calculating the similarity between the actual main frequency feature vector of the target area and the rupture main frequency feature vector under the corresponding moisture content comprises, obtaining the actual main frequency feature vector and the rupture main frequency feature vector under the corresponding moisture content extracted in the test model; The similarity between the actual main frequency feature vector of the target area and the rupture main frequency feature vector under the corresponding moisture content is calculated according to the formula.

[0015] Further, the process of determining the stability level of the high-position landslide based on the similarity comprises, If the hydrological influence level of the target area is the first hydrological influence level, and the similarity is greater than a first similarity threshold, the stability level of the high-position landslide is determined as a high-risk level; If the hydrological influence level of the target area is the second hydrological influence level, and the similarity is greater than a second similarity threshold, the stability level of the high-position landslide is determined as a warning level.

[0016] Further, the process of determining whether to trigger emergency evacuation based on the stability level of the high-position landslide comprises, If the stability level of the high-position landslide in the target area is the high-risk level, a sound and light alarm device is started, and an evacuation instruction is sent; If the stability level of the high-position landslide in the target area is the warning level, the frequency of collecting microseismic signals is determined as a higher frequency, and the actual microseismic signals of the target area are monitored.

[0017] Further, an intelligent terminal for high-order landslide stability analysis based on microseismic signal recognition is also included, which comprises a data acquisition module, a data analysis module, a grade evaluation module and a warning module.

[0018] Compared with the prior art, the present application has the beneficial effect that the present application provides a high-order landslide stability analysis method based on microseismic signal recognition. By performing tensile and shear failure experiments on rock samples under different water contents, corresponding microseismic signals are obtained and a test feature database is constructed, which can accurately capture the fracture characteristics of rock mass under different hydrological environments, making up for the defects of traditional methods that ignore the influence of hydrological conditions on rock mass fracture. At the same time, the characteristic value is calculated combined with historical hydrological characteristic parameters and divided into grades to determine the microseismic signal collection frequency, realizing the dynamic adjustment of the monitoring frequency, making the monitoring work more targeted, ensuring the monitoring effect while avoiding resource waste. In addition, by collecting and analyzing the adjusted microseismic signals multiple times, a closed-loop monitoring and evaluation system is formed, which can timely grasp the change of high-order landslide stability and provide a scientific basis for subsequent decision-making, significantly improving the precision of high-order landslide warning.

[0019] Especially, the present application can present the complex microseismic signal in the form of intuitive image by generating the energy distribution visual image in the "time-frequency" two-dimensional coordinate system of the test microseismic signal under each water content, which is convenient for in-depth analysis of the signal. The main frequency feature vector of the fracture is extracted based on the energy distribution visual image, which comprehensively reflects the essential characteristics of the microseismic signal when the rock mass is fractured, including the energy peak frequency, the ratio of the main frequency energy to the total energy, and the frequency bandwidth. This method not only improves the accuracy and comprehensiveness of feature extraction, but also provides a reliable basis for comparative analysis of rock mass fracture characteristics under different water contents. Compared with the traditional method of simply analyzing the microseismic signal, this process can more accurately capture the subtle changes of rock mass fracture, thereby providing a more solid foundation for subsequent high-order landslide stability analysis, and helping to improve the reliability and effectiveness of the entire analysis method.

[0020] Especially, the application converts these dispersed parameters into a comprehensive hydrological characteristic representation value by processing the soil moisture content, groundwater level change range and rainfall in the historical period. Firstly, the ratio calculation is performed on each parameter and the corresponding predetermined threshold value, realizing the standardization processing of each parameter, so that parameters of different orders of magnitude have comparability. Then, the ratio values of the three parameters are added to obtain a comprehensive representation value, which can comprehensively reflect the overall situation of the hydrological characteristics of the target region, avoiding the one-sidedness of single parameter analysis. This method considers the comprehensive influence of various hydrological factors on high-position landslide, so that the calculated hydrological characteristic representation value can more truly reflect the hydrological conditions of the target region, providing scientific and reasonable data support for subsequent hydrological influence grade division, which helps to improve the accuracy of high-position landslide stability analysis.

[0021] Especially, the application divides the hydrological influence grade of the target region by the hydrological characteristic representation value, which can clearly and definitely define the hydrological conditions of the target region. By setting a predetermined hydrological characteristic representation value threshold, the hydrological influence grade of the target region is divided into first and second grades, so that the complex hydrological conditions become simple and easy to understand, which is convenient for subsequent corresponding measures. This division method can quickly judge the influence degree of the hydrological conditions of the target region on the stability of high-position landslide. When the hydrological characteristic representation value is greater than the threshold value, it indicates that the hydrological conditions have a greater influence on the high-position landslide, and more stringent monitoring and prevention measures need to be taken. Otherwise, the influence is relatively small, and the monitoring strategy can be adjusted appropriately. This grade division method provides a direct basis for determining the frequency of microseismic signal collection, which helps to improve the pertinence and effectiveness of high-position landslide monitoring, thereby improving the practicality of the entire stability analysis method.

[0022] Especially, the application can objectively and quantitatively measure the similarity between the two by calculating the similarity of the actual main frequency feature vector and the corresponding main frequency feature vector under the water content using the cosine similarity calculation formula. The cosine similarity reflects the similarity of two vectors by calculating the angle between them in space. The greater the value, the more similar the two. This calculation method can consider the influence of each parameter in the feature vector, avoiding the limitations of single parameter comparison, making the similarity calculation result more scientific and reliable. Through this quantitative similarity value, the proximity of the current target region rock mass condition and the rupture state can be more accurately judged, providing a clear numerical basis for the determination of high-position landslide stability grade, which helps to improve the objectivity and accuracy of the determination result, reduces the error of human judgment, and thus improves the credibility of the entire high-position landslide stability analysis method.

[0023] Especially, the application determines the high-position landslide stability grade by similarity and hydrological influence grade, fully considers the comprehensive influence of the rock mass itself condition and external hydrological condition, and makes the determination result more comprehensive and reasonable. Different similarity thresholds are used for determination under different hydrological influence grades, which reflects the important role of hydrological condition on the high-position landslide stability. When the hydrological influence grade is the first grade, the external hydrological condition is relatively high, if the similarity is greater than the first similarity threshold, the high-risk grade is determined, and the high-risk state can be identified in time; when the hydrological influence grade is the second grade, the external hydrological condition is relatively stable, if the similarity is greater than the second similarity threshold, the warning grade is determined, and the prevention preparation can be made in advance. This determination method avoids the one-sidedness of determining the grade according to a single factor, and improves the accuracy of the high-position landslide stability grade determination. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 A step flow chart of a high-position landslide stability analysis method based on microseismic signal recognition of an embodiment of the application is shown in the figure, and the application provides a high-position landslide stability analysis method based on microseismic signal recognition, which comprises the following steps: Figure 2 A step flow chart of acquiring a broken main frequency feature vector under each water content based on the test microseismic signal under each water content of the embodiment of the application is shown in the figure; Figure 3 A step flow chart of calculating a hydrological feature representation value based on the hydrological feature parameters of the target area in the historical period of the embodiment of the application is shown in the figure; Figure 4 A determination logic diagram of dividing the hydrological influence grade of the target area based on the hydrological feature representation value of the embodiment of the application. DETAILED DESCRIPTION

[0025] In order to make the purpose and advantages of the application more clear and understandable, the application will be further described below in combination with embodiments; it should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the protection scope of the application.

[0026] The preferred embodiments of the application will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the application, and are not used to limit the protection scope of the application.

[0027] Please refer to Figure 1 The figure is a step flow chart of a high-position landslide stability analysis method based on microseismic signal recognition of an embodiment of the application, and the application provides a high-position landslide stability analysis method based on microseismic signal recognition, which comprises the following steps: Collect the slope rock sample of the target area to perform a damage verification experiment under different water content conditions, and collect the test microseismic signal under each water content by using an acoustic emission sensor; construct a test feature database based on the test microseismic signals at the water contents, and obtain a rupture main frequency feature vector at each water content; collect hydrological feature parameters of the target region in a historical period, calculate a hydrological feature representation value based on the hydrological feature parameters of the target region in the historical period; divide a hydrological influence level of the target region based on the hydrological feature representation value, and determine a frequency of collecting microseismic signals based on the hydrological influence level; collect actual microseismic signals of the target region based on the frequency of collecting microseismic signals, construct a field feature database based on the actual microseismic signals and soil water content, and obtain an actual main frequency feature vector; calculate a similarity between the actual main frequency feature vector of the target region and a rupture main frequency feature vector corresponding to the water content, and determine a high-position landslide stability level based on the similarity; determine whether to trigger emergency evacuation and whether to start a drainage device and a grouting device based on the high-position landslide stability level; after completing drainage and grouting adjustment, collect actual microseismic signals of the adjusted target region, obtain a corresponding actual main frequency feature vector, a corresponding updated similarity, and determine a corresponding high-position landslide stability level based on the similarity; The hydrological feature parameters include soil water content, underground water level change amplitude, and rainfall.

[0028] It can be understood that the rupture mechanism of the damage verification experiment includes tensile failure and shear failure.

[0029] It can be understood that the collection process of the test microseismic signals includes obtaining acoustic emission signals of the rock sample in the tensile failure experiment and the shear failure experiment through an acoustic emission sensor; and the actual microseismic signals are collected in the target region by microseismic monitoring equipment.

[0030] By performing tensile and shear failure experiments on rock samples under different water contents, corresponding microseismic signals are obtained to construct a test feature database, which can accurately capture the rupture characteristics of rock mass under different hydrological environments, and makes up for the defects of traditional methods that ignore the influence of hydrological conditions on rock mass failure. At the same time, the representation value is calculated combined with historical hydrological feature parameters, and the level is divided, so as to determine the frequency of collecting microseismic signals, realize dynamic adjustment of the monitoring frequency, make the monitoring work more targeted, ensure the monitoring effect, and avoid waste of resources. In addition, by collecting and analyzing the adjusted microseismic signals for multiple times, a closed-loop monitoring and evaluation system is formed, which can timely grasp the change of the high-position landslide stability, provides a scientific basis for subsequent decision-making, and significantly improves the high-position landslide early warning accuracy.

[0031] Please refer to Figure 2As shown, it is the step flow chart of the embodiment of the application for constructing a test feature database based on the test microseismic signals under each water content, and acquiring the rupture main frequency feature vector under each water content, the process of the embodiment of the application for constructing a test feature database based on the test microseismic signals under each water content, and acquiring the rupture main frequency feature vector under each water content includes: The test microseismic signals under each water content obtained are generated into energy distribution visual images in a "time-frequency" two-dimensional coordinate system; Based on the energy distribution visual images, the rupture main frequency feature vector under each water content is acquired; The rupture main frequency feature vector includes an energy peak frequency, a ratio of main frequency energy to total energy, and a frequency band width.

[0032] In this embodiment, the test microseismic signals under each water content obtained are removed from environmental noise, and effective signal segments are retained, energy distribution visual images in a "time-frequency" two-dimensional coordinate system are generated, the energy distribution of the microseismic signals in the time and frequency dimensions is intuitively displayed, the energy peak frequency, the ratio of main frequency energy to total energy, and the frequency band width are extracted by using a peak detection algorithm, and the rupture main frequency feature vector under each water content is constructed, thereby providing a standardized data format for subsequent analysis.

[0033] By generating the test microseismic signals under each water content into energy distribution visual images in a "time-frequency" two-dimensional coordinate system, the complex microseismic signals can be presented in the form of intuitive images, and the signals are facilitated to be analyzed in depth. The rupture main frequency feature vector is extracted based on the energy distribution visual images, and the energy peak frequency, the ratio of main frequency energy to total energy, and the frequency band width contained in the rupture main frequency feature vector comprehensively reflect the essential characteristics of the microseismic signals when the rock mass is broken. This method not only improves the accuracy and comprehensiveness of feature extraction, but also provides a reliable basis for comparative analysis of rock mass rupture characteristics under different water contents. Compared with the traditional method of simply analyzing the microseismic signals, this process can more accurately capture the subtle changes when the rock mass is broken, thereby providing a more solid foundation for subsequent high landslide stability analysis, and helping to improve the reliability and effectiveness of the entire analysis method.

[0034] Please refer to Figure 3 As shown, it is the step flow chart of the embodiment of the application for calculating a hydrological feature representation value based on the hydrological feature parameters of the target region in the historical period, the process of the embodiment of the application for calculating a hydrological feature representation value based on the hydrological feature parameters of the target region in the historical period includes: The process of calculating a hydrological feature representation value based on the hydrological feature parameters of the target region in the historical period includes, The soil water content, the groundwater level change amplitude, and the rainfall of the target region in the historical period are acquired. a ratio of the soil moisture content in a single period to a predetermined soil moisture content threshold value is determined as a first hydrological characteristic parameter; a ratio of the groundwater level change amplitude in a single period to a predetermined groundwater level change amplitude threshold value is determined as a second hydrological characteristic parameter; a ratio of the rainfall in a single period to a predetermined rainfall threshold value is determined as a third hydrological characteristic parameter; a sum of the first hydrological characteristic parameter, the second hydrological characteristic parameter and the third hydrological characteristic parameter is determined as a hydrological characteristic representation value.

[0035] In the embodiment, the soil moisture content threshold value is obtained in advance, the soil moisture content in a rainy season in a historical period of twelve months is obtained, the mean value is calculated, and the soil moisture content threshold value is set to 75% of the mean value. The groundwater level change amplitude threshold value is obtained in advance, the daily groundwater level change amplitude in a historical period of three months is obtained, the mean value is calculated, and the groundwater level change amplitude threshold value is set to twice the mean value. The rainfall threshold value is obtained in advance, the rainfall in a rainy season in a historical period of twelve months is obtained, the mean value is calculated, and the rainfall threshold value is set to 75% of the mean value. A single period is set to 24 hours.

[0036] By processing the multi-dimensional hydrological characteristic parameters of soil moisture content, groundwater level change amplitude and rainfall in a historical period, these dispersed parameters are converted into a comprehensive hydrological characteristic representation value. First, the ratio of each parameter to the corresponding predetermined threshold value is calculated to realize the standardization processing of each parameter, so that parameters of different orders of magnitude have comparability. Then, the ratio of the three parameters is added to obtain a comprehensive representation value, which can comprehensively reflect the overall situation of the hydrological characteristics of the target region and avoid the one-sidedness of single parameter analysis. This method considers the comprehensive influence of various hydrological factors on high-position landslide, so that the calculated hydrological characteristic representation value can more truly reflect the hydrological conditions of the target region, provides scientific and reasonable data support for subsequent hydrological influence grade division, and helps to improve the accuracy of high-position landslide stability analysis.

[0037] Please refer to Figure 4 Fig. 1 is a determination logic diagram for dividing the hydrological influence grade of a target region based on the hydrological characteristic representation value according to an embodiment of the present application, and the process of dividing the hydrological influence grade of a target region based on the hydrological characteristic representation value according to the embodiment of the present application includes: The process of dividing the hydrological influence grade of a target region based on the hydrological characteristic representation value includes, If the hydrological characteristic representation value of the target region is greater than a predetermined hydrological characteristic representation value threshold value, it is determined that the hydrological influence grade of the target region is a first hydrological influence grade; If the hydrological characteristic value of the target region is less than or equal to a predetermined hydrological characteristic value threshold, it is determined that the hydrological influence level of the target region is a second hydrological influence level.

[0038] In the embodiment, the hydrological characteristic value threshold is obtained in advance, and is in the interval [2.85, 2.95].

[0039] By dividing the hydrological influence level of the target region according to the hydrological characteristic value, the hydrological condition of the target region can be clearly and definitely defined. By setting the predetermined hydrological characteristic value threshold, the hydrological influence level of the target region is divided into two levels, namely the first and second levels, so that the complex hydrological condition becomes simple and easy to understand, and it is convenient to take corresponding measures subsequently. This division method can quickly determine the influence degree of the hydrological condition of the target region on the stability of the high-position landslide. When the hydrological characteristic value is greater than the threshold, it indicates that the influence of the hydrological condition on the high-position landslide is greater, and more stringent monitoring and prevention measures need to be taken. Otherwise, the influence is relatively small, and the monitoring strategy can be adjusted appropriately. This level division method provides a direct basis for determining the frequency of collecting microseismic signals, which helps to improve the pertinence and effectiveness of high-position landslide monitoring, and thus improves the practicability of the entire stability analysis method.

[0040] Specifically, the process of determining the frequency of collecting microseismic signals based on the hydrological influence level of the target region includes, If the hydrological influence level of the target region is the first hydrological influence level, the frequency of collecting microseismic signals is determined to be a higher frequency; If the hydrological influence level of the target region is the second hydrological influence level, the frequency of collecting microseismic signals is determined to be a lower frequency.

[0041] In the embodiment, in the first hydrological influence level, a higher frequency is used to collect microseismic signals, 500Hz sampling is used, which is used for high-risk hydrological period and needs to capture high-frequency damage signals. In the second hydrological influence level, a lower frequency is used to collect microseismic signals, 100Hz sampling is used, which is used for dry or low hydrological risk period, which meets the basic monitoring needs and takes into account the endurance of the monitoring equipment.

[0042] Specifically, the process of constructing a field feature database based on the actual microseismic signals of the target region and the soil moisture content and obtaining an actual main frequency feature vector includes, The actual microseismic signals of the target region and the soil moisture content are used to generate an energy distribution visualization image in a time-frequency two-dimensional coordinate system; Based on the energy distribution visualization image, an actual main frequency feature vector is obtained; The actual main frequency feature vector includes an energy peak frequency, a ratio of main frequency energy to total energy, and a frequency band width.

[0043] In the embodiment, the actual microseismic signal of the target area obtained is removed from the environmental noise, the effective signal segment is retained, an energy distribution visualization image in a "time-frequency" two-dimensional coordinate system is generated, the energy distribution of the actual microseismic signal under different soil water content conditions in the time and frequency dimensions is exhibited, the frequency interval with concentrated energy is determined, a peak detection algorithm is adopted, the energy peak frequency, the ratio of the main frequency energy to the total energy, and the frequency band width are extracted, and an actual main frequency feature vector is constructed. The actual main frequency feature vector represents the current actual rock mass microseismic characteristics of the target area, and is used for comparative analysis with the feature vector in the test model.

[0044] Specifically, the process of calculating the similarity between the actual main frequency feature vector of the target area and the main frequency feature vector of the rock failure under the corresponding water content in the test model includes, obtaining the actual main frequency feature vector and the main frequency feature vector of the rock failure under the corresponding water content extracted in the test model; The calculation formula calculates the similarity between the actual main frequency feature vector of the target area and the main frequency feature vector of the rock failure under the corresponding water content.

[0045] It can be understood that the cosine similarity calculation formula is a prior art, which measures the similarity by comparing the directions of two vectors. When calculating, first calculate the sum of the products of the corresponding elements of the two vectors, then calculate the square roots of the squares of the elements of the two vectors respectively, and finally divide the product of the former two by the product of the latter two. The closer the result is to 1, the more consistent the directions of the two vectors are, and the higher the similarity is.

[0046] By using the cosine similarity calculation formula to calculate the similarity between the actual main frequency feature vector and the main frequency feature vector of the rock failure under the corresponding water content, the similarity between the two can be objectively and quantitatively measured. The cosine similarity reflects the similarity of the two vectors by calculating the angle between them in space. The larger the value is, the more similar the two are. This calculation method can consider the influence of each parameter in the feature vector, avoiding the limitations of single parameter comparison, making the calculation result of similarity more scientific and reliable. Through this quantitative similarity value, the proximity of the current rock mass condition and the failure state of the target area can be more accurately judged, providing a clear numerical basis for the determination of the stability level of high-position landslide, which helps to improve the objectivity and accuracy of the determination result, reduces the error of human judgment, and thus improves the credibility of the whole high-position landslide stability analysis method.

[0047] Specifically, the process of determining the stability level of high-position landslide based on the similarity includes, In the case that the hydrological influence level of the target area is the first hydrological influence level, if the similarity is greater than a first similarity threshold, it is determined that the stability level of high-position landslide is the high-risk level; In the case that the hydrological influence level of the target area is the second hydrological influence level, if the similarity is greater than the second similarity threshold, the high-position landslide stability level is determined as the warning level.

[0048] In the embodiment, the first hydrological influence level is the higher risk level, at this time, the water level disturbance causes the natural frequency to shift, the first similarity threshold is set to 0.75, the second hydrological influence level is the lower risk level, and the breaking feature needs to be strictly matched under the low hydrological disturbance, and the second similarity threshold is set to 0.85.

[0049] The high-position landslide stability level is determined by the similarity and the hydrological influence level, the comprehensive influence of the rock mass itself and the external hydrological condition is fully considered, the determination result is more comprehensive and reasonable. In different hydrological influence levels, different similarity thresholds are used for determination, which reflects the important role of the hydrological condition on the high-position landslide stability. When the hydrological influence level is the first level, it is indicated that the external hydrological condition risk is higher, at this time, if the similarity is greater than the first similarity threshold, it is determined as the high-risk level, which can identify the high-risk state in time; when the hydrological influence level is the second level, the external hydrological condition is relatively stable, if the similarity is greater than the second similarity threshold, it is determined as the warning level, which is convenient for making preparation in advance. This determination method avoids the one-sidedness of determining the level according to a single factor, and improves the accuracy of the high-position landslide stability level determination.

[0050] Specifically, based on the high-position landslide stability level, a process of determining whether to trigger an emergency evacuation includes, In the case that the high-position landslide stability level of the target area is the high-risk level, a sound and light alarm device is started, and an evacuation instruction is sent. In the case that the high-position landslide stability level of the target area is the warning level, a frequency of collecting the microseismic signal is determined as a higher frequency, and the actual microseismic signal of the target area is monitored.

[0051] Specifically, it further includes an intelligent terminal of a high-position landslide stability analysis method based on microseismic signal recognition, the intelligent terminal includes a data acquisition module, a data analysis module, a level evaluation module and a warning module.

[0052] So far, the technical scheme of the present application has been described in combination with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical scheme after the changes or replacements will fall within the protection scope of the present application.

Claims

1. A high-order landslide stability analysis method based on microseismic signal recognition, characterized in that, The method comprises the following steps: Collecting slope rock samples of a target area to conduct failure verification experiments under different water content conditions, and collecting test microseismic signals under each water content condition using acoustic emission sensors; Based on the test microseismic signals under each water content condition, a test feature database is constructed, and a rupture main frequency feature vector under each water content condition is obtained; Collecting hydrological characteristic parameters of the target area in a historical period, and calculating a hydrological feature representation value based on the hydrological characteristic parameters of the target area in the historical period; Based on the hydrological feature representation value, the hydrological influence level of the target area is divided, and the frequency of collecting microseismic signals is determined based on the hydrological influence level; Based on the frequency of collecting microseismic signals, actual microseismic signals of the target area are collected, a field feature database is constructed based on the actual microseismic signals and soil water content, and an actual main frequency feature vector is obtained; The similarity between the actual main frequency feature vector of the target area and the rupture main frequency feature vector corresponding to the water content is calculated, and the high-position landslide stability level is determined based on the similarity; Based on the high-position landslide stability level, it is determined whether to trigger emergency evacuation, and whether to start the drainage equipment and grouting equipment; After completing drainage and grouting adjustment, the actual microseismic signals of the adjusted target area are collected, the corresponding actual main frequency feature vector is obtained, the corresponding updated similarity is obtained, and the corresponding high-position landslide stability level is determined based on the similarity; The hydrological characteristic parameters include soil water content, groundwater level change amplitude, and rainfall.

2. The microseismic signal recognition-based high-position landslide stability analysis method according to claim 1, characterized in that, The process of constructing a test feature database based on the test microseismic signals under each water content condition and obtaining a rupture main frequency feature vector under each water content condition comprises, The test microseismic signals obtained under each water content condition are generated into an energy distribution visual image in a "time-frequency" two-dimensional coordinate system; Based on the energy distribution visual image, a rupture main frequency feature vector under each water content condition is obtained; The rupture main frequency feature vector includes an energy peak frequency, a ratio of main frequency energy to total energy, and a frequency band width. 3.The microseismic signal recognition based high-position landslide stability analysis method according to claim 1, characterized in that, The process of calculating a hydrological feature representation value based on the hydrological characteristic parameters of the target area in a historical period comprises, The soil water content, groundwater level change amplitude, and rainfall of the target area in a historical period are obtained; The ratio of soil water content in a single period to a predetermined soil water content threshold value is calculated to determine a first hydrological characteristic parameter; The ratio of groundwater level change amplitude in a single period to a predetermined groundwater level change amplitude threshold value is calculated to determine a second hydrological characteristic parameter; The ratio of rainfall in a single period to a predetermined rainfall threshold value is calculated to determine a third hydrological characteristic parameter; The sum of the first, second, and third hydrological characteristic parameters is determined as the hydrological feature representation value.

4. The microseismic signal recognition-based high-position landslide stability analysis method according to claim 3, characterized in that, The process of dividing the hydrological influence level of the target area based on the hydrological feature representation value comprises, If the hydrological feature representation value of the target area is greater than a predetermined hydrological feature representation threshold value, it is determined that the hydrological influence level of the target area is a first hydrological influence level; If the hydrological feature representation value of the target area is less than or equal to a predetermined hydrological feature representation threshold value, it is determined that the hydrological influence level of the target area is a second hydrological influence level. 5.The microseismic signal recognition based high-position landslide stability analysis method according to claim 1, characterized in that, The process of determining the frequency of collecting microseismic signals based on the hydrological influence level of the target area comprises, if the hydrological influence level of the target area is the first hydrological influence level, determining the frequency of collecting microseismic signals as a higher frequency; if the hydrological influence level of the target area is the second hydrological influence level, determining the frequency of collecting microseismic signals as a lower frequency.

6. The microseismic signal recognition-based high-position landslide stability analysis method according to claim 5, characterized in that, The process of constructing a field feature database based on the actual microseismic signals of the target area and the soil moisture content and obtaining an actual main frequency feature vector comprises, generating an energy distribution visualization image in a time-frequency two-dimensional coordinate system based on the obtained actual microseismic signals of the target area and the soil moisture content; obtaining an actual main frequency feature vector based on the energy distribution visualization image; wherein the actual main frequency feature vector comprises an energy peak frequency, a ratio of main frequency energy to total energy, and a frequency band width.

7. The microseismic signal recognition-based high-position landslide stability analysis method according to claim 6, characterized in that, The process of calculating the similarity between the actual main frequency feature vector of the target area and the rupture main frequency feature vector under the corresponding moisture content comprises, obtaining the actual main frequency feature vector and the rupture main frequency feature vector under the corresponding moisture content extracted in the test model; calculating the similarity between the actual main frequency feature vector of the target area and the rupture main frequency feature vector under the corresponding moisture content using the formula. 8.The microseismic signal recognition based high-position landslide stability analysis method according to claim 7, characterized in that, The process of determining the high landslide stability level based on the similarity comprises, if the hydrological influence level of the target area is the first hydrological influence level, and the similarity is greater than a first similarity threshold, determining the high landslide stability level as a high-risk level; if the hydrological influence level of the target area is the second hydrological influence level, and the similarity is greater than a second similarity threshold, determining the high landslide stability level as a warning level. 9.The high-position landslide stability analysis method based on microseismic signal recognition of claim 8, characterized in that, The process of determining whether to trigger emergency evacuation based on the high landslide stability level comprises, if the high landslide stability level of the target area is the high-risk level, starting an audible and visual alarm device and sending an evacuation instruction; if the high landslide stability level of the target area is the warning level, determining the frequency of collecting microseismic signals as a higher frequency and monitoring the actual microseismic signals of the target area. 10.The high-position landslide stability analysis method based on microseismic signal recognition of claim 9, wherein, The intelligent terminal used in the high landslide stability analysis method comprises a data acquisition module, a data analysis module, a level evaluation module, and a warning module.

Images