Slope settlement monitoring method and system

Abstract

The application provides a kind of side slope subsidence monitoring method and system, it is related to side slope monitoring technical field, first acquire the multi-source cause information of side slope area, including side slope surface soil disturbance, underground rock-soil stress change, surrounding water body penetration and vegetation cover change information, then based on multi-source cause information constructs side slope subsidence cause conduction chain, embodies the subsidence influence conduction relationship between causes, again according to side slope subsidence cause conduction chain generates cause conduction tracking sequence, records conduction order and state change information, then according to cause conduction tracking sequence determines the key conduction node of side slope subsidence, finally based on the position and state change information of key conduction node, generates the monitoring resource directional deployment scheme containing monitoring equipment deployment position and monitoring frequency setting, can thus comprehensively, accurately monitor side slope subsidence, improves monitoring efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of slope monitoring technology, and more specifically, to a slope settlement monitoring method and system. Background Technology

[0002] In the fields of geological engineering and environmental monitoring, slope settlement has always been a major concern. The stability of slopes directly affects the safety of surrounding infrastructure, residents' lives and property, and the ecological environment. With the acceleration of urbanization and the continuous development of various engineering projects, the environment in which slopes are located is becoming increasingly complex and is affected by a variety of factors.

[0003] Traditional slope settlement monitoring methods often focus on a single monitoring indicator, such as displacement monitoring of the slope surface. However, slope settlement is a complex dynamic process influenced by a combination of factors. Disturbance of the surface soil may be caused by human activities (such as construction and excavation) or natural factors (such as earthquakes and rainfall erosion); changes in stress in the underground soil and rock layers are affected by geological tectonic movements, changes in groundwater levels, and other factors; infiltration of surrounding water bodies can alter the water content and mechanical properties of the slope soil; and changes in vegetation cover can affect the ecological stability and soil and water conservation capacity of the slope.

[0004] Existing monitoring technologies are insufficient to comprehensively and systematically consider these multiple causes and their interrelationships, making it impossible to accurately predict the development trend of slope settlement. This results in significant limitations in slope settlement monitoring and early warning, making it difficult to meet the high-precision and comprehensive requirements for slope safety monitoring in actual engineering projects. Summary of the Invention

[0005] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a slope settlement monitoring method, the method comprising:

[0006] Acquire multi-source inducing information for the slope area, including information on disturbance of the surface soil of the slope, information on stress change of the underground soil and rock layer, information on infiltration of the surrounding water body, and information on vegetation cover change.

[0007] Based on the multi-source causal information, a slope settlement causal transmission chain is constructed. The slope settlement causal transmission chain uses each causal information as a chain node, and the connection relationship between the chain nodes reflects the transmission relationship of settlement influence between the causal factors.

[0008] Based on the slope settlement induction cause transmission chain, an induction cause transmission tracing sequence is generated. The induction cause transmission tracing sequence records the transmission order between chain nodes and the state change information during the transmission process.

[0009] Based on the aforementioned causal transmission tracing sequence, the key transmission nodes of slope settlement are determined, and the key transmission nodes are the chain nodes that have the highest influence on the settlement trend;

[0010] Based on the position and state change information of the key conduction node, a monitoring resource directional deployment scheme is generated, and the monitoring resource directional deployment scheme includes a monitoring device deployment position and a monitoring frequency setting for the key conduction node.

[0011] In still another aspect, the embodiment of the present application further provides a slope settlement monitoring system, characterized by comprising:

[0012] a processor; a machine readable storage medium for storing machine executable instructions of the processor; wherein the processor is configured to execute the above-mentioned slope settlement monitoring method by executing the machine executable instructions.

[0013] In still another aspect, the embodiment of the present application further provides a computer program product, which comprises machine executable instructions stored in a computer readable storage medium, and a processor of a computer device reads the machine executable instructions from the computer readable storage medium, and the processor executes the machine executable instructions, so that the computer device executes the above-mentioned slope settlement monitoring method.

[0014] Based on the above aspects, by acquiring multi-source cause information of the slope area, various factors affecting the slope settlement can be known, including slope surface soil disturbance, underground rock and soil stress change, surrounding water body penetration, and vegetation coverage change, and then a slope settlement cause conduction chain is constructed based on the multi-source cause information, each cause information is taken as a chain node and the settlement influence conduction relationship therebetween is determined, a cause conduction tracking sequence is generated according to the slope settlement cause conduction chain, the conduction order and state change information between the chain nodes are recorded, the conduction process of the cause can be dynamically tracked, the change of the cause in the conduction process can be found in time, the key conduction node of the slope settlement is determined according to the cause conduction tracking sequence, the chain node with the highest influence degree on the settlement trend is accurately positioned, a monitoring resource directional deployment scheme is generated based on the position and state change information of the key conduction node, and the monitoring resource directional deployment scheme includes a monitoring device deployment position and a monitoring frequency setting for the key conduction node, so that the accurate placement and efficient use of the monitoring resource are realized, the monitoring data of the key node can be obtained in time and accurately, and the efficiency and accuracy of the slope settlement monitoring are greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is an execution flow schematic diagram of the slope settlement monitoring method provided by the embodiment of the present application.

[0016] Figure 2 is a schematic diagram of exemplary hardware and software components of the slope settlement monitoring system provided by the embodiment of the present application. DETAILED DESCRIPTION

[0017] The present application will be described in detail below with reference to the accompanying drawings, Figure 1 is a flowchart of a slope settlement monitoring method provided by an embodiment of the present application, and the slope settlement monitoring method will be described in detail below.

[0018] Step S110: Obtain multi-source cause information of the slope area, which includes slope surface soil disturbance information, underground rock and soil stress change information, surrounding water body penetration information, and vegetation coverage change information.

[0019] In this embodiment, a mountainous area highway slope is taken as the monitoring object to carry out slope settlement monitoring work. When obtaining multi-source cause information, for the slope surface soil disturbance information, displacement monitoring equipment can be arranged in different areas of the slope surface, such as installing high-precision GNSS (Global Navigation Satellite System) receivers at the top, middle and foot of the slope, and the sampling frequency of the receiver can be set according to the monitoring requirements, which can generally be set to collect data once an hour. The collected displacement data includes horizontal displacement and vertical displacement, which constitute the core content of the slope surface soil disturbance information. For the underground rock and soil stress change information, stress sensors need to be buried in different rock and soil layers under the slope, such as in different rock and soil layers such as clay layer, sand layer and weathered rock layer, and vibration string type stress sensors are buried at certain intervals (such as every certain distance). The stress data collected by the sensors are transmitted to the data acquisition terminal in real time to form the underground rock and soil stress change information. The surrounding water body penetration information can be obtained by installing water level sensors and penetration pressure sensors in the surface water area (such as drainage ditches and water pits) around the slope and the seepage monitoring hole inside the slope. The water level sensor collects water level change data, and the penetration pressure sensor collects penetration pressure change data, which together constitute the surrounding water body penetration information. The vegetation coverage change information can be obtained by regularly (such as once a month) obtaining high-resolution images of the slope area by unmanned aerial vehicle aerial photography combined with satellite remote sensing, and then processing the images by using image recognition algorithm to analyze the coverage area of the vegetation and the change of the vegetation type, so as to obtain the vegetation coverage change information.

[0020] Step S120: Construct a slope settlement cause conduction chain based on the multi-source cause information, wherein each cause information is taken as a chain node, and the connection relationship between the chain nodes reflects the settlement influence conduction relationship between the causes.

[0021] Step S121: classifying the multi-source cause information according to the action level of affecting the slope settlement, to obtain surface action cause information, underground action cause information, hydrological action cause information and biological action cause information, the surface action cause information corresponds to the slope surface soil disturbance information, the underground action cause information corresponds to the underground rock and soil stress change information, the hydrological action cause information corresponds to the surrounding water body penetration information, and the biological action cause information corresponds to the vegetation coverage change information.

[0022] In the present embodiment, the multi-source cause information obtained above is classified according to the action level. Among them, the slope surface soil disturbance information (such as the slope surface displacement data collected by the GNSS receiver) belongs to the surface action cause information, because it directly acts on the surface soil of the slope and affects the stability of the surface soil; the underground rock and soil stress change information (such as the stress data collected by the vibrating wire stress sensor) belongs to the underground action cause information, which acts on the underground rock and soil and affects the overall stability of the slope through the stress transmission of the rock and soil; the surrounding water body penetration information (such as the data collected by the water level sensor and the penetration pressure sensor) belongs to the hydrological action cause information, the penetration of the water body will change the physical properties of the rock and soil, and then affect the slope settlement; the vegetation coverage change information (such as the vegetation coverage change data obtained by image recognition) belongs to the biological action cause information, the root system of the vegetation and the change of the vegetation coverage area will affect the anti-sliding ability of the slope soil.

[0023] Step S122: extracting the state change features in each type of cause information, the state change features including the change starting time, the change duration and the change amplitude features.

[0024] For each type of classified cause information, the state change characteristics are extracted. Taking the surface effect cause information as an example, from the displacement data collected by the GNSS receiver, the starting time of displacement change (i.e. the time when the displacement data starts to deviate from the initial stable value), the change duration (the time period from the start of displacement change to the stabilization of displacement), and the change amplitude feature (the difference between the maximum and minimum values of displacement change) are determined; for the underground effect cause information, from the stress data collected by the vibrating wire stress sensor, the starting time of stress change, the change duration (the time from the start of stress change to stabilization), and the change amplitude feature (the difference between the maximum and minimum values of stress change) are extracted; in the hydrological effect cause information, from the data collected by the water level sensor and the osmotic pressure sensor, the starting time of water level or osmotic pressure change, the change duration (the time from the start of change to stabilization), and the change amplitude feature (the difference between the maximum and minimum values of water level or osmotic pressure change) are determined; in the biological effect cause information, from the vegetation coverage change data obtained by image recognition, the starting time of vegetation coverage change, the change duration (the time from the start of vegetation coverage area change to stabilization), and the change amplitude feature (the difference between the maximum and minimum values of vegetation coverage area change) are extracted.

[0025] Step S123: Obtain the rock-soil layer type distribution data of the slope area, which includes the distribution range of clay layer, the distribution range of sandy soil layer, the distribution range of weathered rock layer, and the distribution range of intact rock layer.

[0026] In this embodiment, the rock-soil layer type distribution data of the slope area is obtained through geological exploration. Specifically, drilling is performed at different positions of the slope (such as drilling points at the top, middle and foot of the slope) using the drilling method, and the drilling depth is determined according to the height of the slope and the distribution of rock-soil layers, which can generally be drilled to a certain depth below the intact rock layer. The rock core taken from the drilling is analyzed to determine the type of rock-soil layer (such as clay, sandy soil, weathered rock, and intact rock), and then a distribution range map of clay layer, sandy soil layer, weathered rock layer, and intact rock layer is drawn according to the drilling data to obtain the rock-soil layer type distribution data.

[0027] Step S124: Adjust the extraction accuracy of the state change characteristics according to the rock-soil layer type distribution data, and improve the extraction accuracy of the change amplitude feature in the distribution range of clay layer, and maintain the basic extraction accuracy in the distribution range of intact rock layer.

[0028] According to the rock-soil layer type distribution data, the extraction accuracy of the state change characteristics is adjusted. For example, in the area where the clay layer is distributed, the compressibility and shear strength characteristics of the clay are more sensitive to the influence on the slope settlement, so when extracting the change amplitude characteristics of the soil disturbance information and stress change information in this area, the collection frequency of the GNSS receiver, the vibrating wire stress sensor and the like is increased (for example, from once every hour to once every half hour), so as to obtain more accurate change amplitude data; in the area where the complete rock layer is distributed, the stability of the complete rock layer is better, and the influence on the slope settlement is relatively small, so the basic collection frequency of the sensor is maintained (for example, once every hour), so as to extract the change amplitude characteristics of the basic accuracy.

[0029] Step S125: comparing the state change characteristics of different categories of cause information, when the change start time of a category of cause information is earlier than the change start time of another category of cause information, and there is an overlapping period in the change duration of the two categories of cause information, it is judged whether the action areas of the two categories of cause information exist spatial intersection according to the rock-soil layer type distribution data.

[0030] Step S1251: the change start time and the change duration are extracted from the state change characteristics of each category of cause information, and the change end time of each category of cause information is calculated, which is the sum of the change start time and the change duration.

[0031] Taking the surface action cause information (slope surface soil disturbance information) and the underground action cause information (underground rock-soil layer stress change information) as examples, the change start time t1 and the change duration d1 are extracted from the state change characteristics of the surface action cause information, and the change end time t1_end=t1+d1 is calculated; the change start time t2 and the change duration d2 are extracted from the state change characteristics of the underground action cause information, and the change end time t2_end=t2+d2 is calculated.

[0032] Step S1252: the coordinate range of the action area corresponding to each category of cause information is extracted, and the coordinate range of the action area is determined according to the collection points and the influence range of the multi-source cause information. The coordinate range of the action area of the surface action cause information corresponds to the surrounding area of the collection points of the surface soil disturbance, and the coordinate range of the action area of the underground action cause information corresponds to the surrounding area of the monitoring hole of the underground rock-soil layer stress change.

[0033] For the surface-acting inducing information, the collection point is the installation position of the GNSS receiver, and the influence range is determined according to the properties of the soil body and the slope of the slope and other factors. For example, for the GNSS receiver at the top of the slope, the coordinate range of the action area can be a circular area with the receiver as the center and a certain distance (such as the distance calculated according to the diffusion angle of the soil body) as the radius. For the underground-acting inducing information, the collection point is the buried position of the vibrating wire stress sensor, and the influence range is determined according to the transmission characteristics of the rock-soil layer. For example, for the stress sensor in the sand layer, the coordinate range of the action area can be a cuboid area with a certain distance (such as the distance calculated according to the stress diffusion coefficient of the sand layer) above, below, left and right of the sensor position.

[0034] Step S1253: Superimpose the coordinate range of the action area of each type of inducing information on the rock-soil layer type distribution data to determine the rock-soil layer types contained in the action area of each type of inducing information and the proportion of each type of rock-soil layer.

[0035] Superimpose the coordinate range of the action area of the surface-acting inducing information (such as the circular action area of the GNSS receiver at the top of the slope) on the rock-soil layer type distribution data, and analyze the rock-soil layer types (such as clay, sand, weathered rock, and intact rock) contained in the action area and the proportion of each type of rock-soil layer through geographic information system (GIS) software. For example, the proportion of the area of clay in the total area of the action area, the proportion of the area of sand, etc. Similarly, superimpose and analyze the coordinate range of the action area of the underground-acting inducing information to determine the rock-soil layer types and proportions in the action area.

[0036] Step S1254: Select any two types of inducing information and label them as the first type of inducing information and the second type of inducing information; compare the change start time of the first type of inducing information with the change start time of the second type of inducing information, and when the change start time of the first type of inducing information is earlier than the change start time of the second type of inducing information, calculate the time difference between the change end time of the first type of inducing information and the change start time of the second type of inducing information. If the time difference is less than zero, it is determined that the change end time of the first type of inducing information is later than the change start time of the second type of inducing information, and the change duration of the two types of inducing information overlaps. At this time, extract the coordinate range of the action area of the first type of inducing information and the coordinate range of the action area of the second type of inducing information, and calculate the intersection area of the two coordinate ranges.

[0037] Take the surface-acting trigger information (first type of trigger information) and the underground-acting trigger information (second type of trigger information) as examples, compare their change starting time t1 and t2, and assume that t1 is earlier than t2. Calculate t1_end-t2, if the time difference is less than zero, it means that t1_end is later than t2, that is, there is an overlapping period between the change duration of the surface-acting trigger information and the change duration of the underground-acting trigger information. At this time, the action area coordinate range of the surface-acting trigger information (such as the circular area of the GNSS receiver on the slope top) and the action area coordinate range of the underground-acting trigger information (such as the cuboid area of the stress sensor in the sand layer) are extracted, and the intersection area of the two coordinate ranges is calculated by GIS software.

[0038] Step S1255: The ratio of the intersection area to the total area of the action area of the first type of trigger information is calculated as the spatial intersection ratio; at the same time, the rock-soil layer type corresponding to the intersection area is checked, and if the proportion of clay or sand layer in the intersection area exceeds the preset proportion, it is determined that there is an effective spatial intersection between the action areas of the two types of trigger information.

[0039] The ratio of the intersection area to the total area of the action area of the surface-acting trigger information is calculated to obtain the spatial intersection ratio. For example, the intersection area is A, the total area of the action area of the surface-acting trigger information is B, and the spatial intersection ratio is A / B. At the same time, the rock-soil layer type corresponding to the intersection area is checked, and if the proportion of clay or sand layer in the intersection area exceeds the preset proportion (such as 50%), it is determined that there is an effective spatial intersection between the action areas of the two types of trigger information.

[0040] Step S1256: If the spatial intersection ratio is greater than zero and there is an effective spatial intersection, mark the first type of trigger information as a conductive initiation trigger information and the second type of trigger information as a conductive reception trigger information; if the spatial intersection ratio is zero or the proportion of effective rock-soil layer in the intersection area does not reach the preset proportion, recheck the overlapping period of the change duration of the two types of trigger information, and extend the overlapping period determination cycle to calculate the time difference between the change end time and the change start time again, and rejudge the overlapping situation.

[0041] If the space intersection ratio is greater than zero and the proportion of clay or sand layer in the intersection area exceeds the preset proportion, it indicates that the effective space intersection exists between the action area of the surface action inducing information (first type of inducing information) and the action area of the underground action inducing information (second type of inducing information), and the overlapping period exists in the change duration of the two types of inducing information. Therefore, the surface action inducing information is marked as the transmission initiation inducing information, and the underground action inducing information is marked as the transmission receiving inducing information. If the space intersection ratio is zero or the proportion of effective rock-soil layer in the intersection area does not reach the preset proportion, the overlapping period of the change duration of the two types of inducing information is rechecked, the overlapping period determination period is extended (for example, from the original hour unit to day unit), the time difference between the change end time and the change start time is calculated again, and the overlapping situation is rejudged.

[0042] Step S1257: When the change start time of the first type of inducing information is later than the change start time of the second type of inducing information, the steps of calculating the time difference, the space intersection ratio and the effective rock-soil layer ratio are repeated. If there is an overlapping period and the action area has an effective space intersection, the second type of inducing information is marked as the transmission initiation inducing information, and the first type of inducing information is marked as the transmission receiving inducing information.

[0043] Suppose the change start time t2 of the selected first type of inducing information (such as underground action inducing information) is later than the change start time t1 of the second type of inducing information (such as surface action inducing information), the operations of steps S1254 to S1256 are repeated, and the time difference (t2-t1), the space intersection ratio and the effective rock-soil layer ratio are calculated. If there is an overlapping period and the action area has an effective space intersection, the second type of inducing information (surface action inducing information) is marked as the transmission initiation inducing information, and the first type of inducing information (underground action inducing information) is marked as the transmission receiving inducing information.

[0044] Step S1258: The above operations are repeated for any combination of two types of inducing information, and the rock-soil layer type data and the space intersection data in each judgment process are recorded to form a transmission relationship judgment record, which is convenient for subsequent verification of the rationality of the transmission relationship.

[0045] For all possible combinations of the four types of inducing information (such as surface and hydrology, surface and biology, underground and hydrology, underground and biology, hydrology and biology), the operations of steps S1254 to S1257 are repeated, and the rock-soil layer type data (such as the rock-soil layer type and proportion in the action area), the space intersection data (such as the intersection area, the space intersection ratio) and the like in each judgment process are recorded to form a transmission relationship judgment record, which is convenient for subsequent verification of the rationality of the transmission relationship.

[0046] Step S1259: After the transmission initiation and reception marking between all categories of cause information is completed, the marking results are summarized to form a preliminary transmission initiation and reception relationship table, which includes transmission initiation cause information identification, transmission reception cause information identification, overlapping period data, and spatial intersection data.

[0047] After the transmission initiation and reception marking between all categories of cause information is completed, the marking results are summarized. For example, the transmission initiation cause information identification is surface action cause information, the transmission reception cause information identification is underground action cause information, the overlapping period data is the time period from t1 to t2_end, and the spatial intersection data is intersection area, spatial intersection proportion, and effective rock layer proportion, etc. The above information is sorted into a preliminary transmission initiation and reception relationship table.

[0048] Step S126: The connection frequency between the same transmission initiation cause information and transmission reception cause information in the preliminary transmission chain structure is counted, and the statistical weight of the connection frequency is adjusted in combination with the influence degree of the rock and soil layer type on the transmission, and the statistical weight of the connection frequency in the clay layer distribution area is greater than that in the complete rock layer distribution area.

[0049] In the preliminary transmission chain structure, the connection frequency between the same transmission initiation cause information and transmission reception cause information is counted. For example, the surface action cause information is taken as the transmission initiation cause information, and the underground action cause information is taken as the transmission reception cause information, and the connection frequency between them is counted. Then, the statistical weight of the connection frequency is adjusted in combination with the influence degree of the rock and soil layer type on the transmission. Since the compressibility and shear strength characteristics of clay layer have a greater influence on the transmission, the statistical weight of the connection frequency in the clay layer distribution area is greater than that in the complete rock layer distribution area. For example, the connection frequency weight in the clay layer distribution area is set to 2, and the connection frequency weight in the complete rock layer distribution area is set to 1.

[0050] Step S127: The transmission strength between the chain nodes is determined according to the adjusted connection frequency, and the connection frequency value is large, and the transmission strength value is large.

[0051] According to the adjusted connection frequency, the transmission strength between the chain nodes is calculated. For example, the adjusted connection frequency between the surface action cause information and the underground action cause information is N (N is the value after the connection frequency is multiplied by the weight), and the transmission strength is proportional to N, and the larger N is, the larger the transmission strength value is.

[0052] Step S128: In the preliminary transmission chain structure, the transmission strength between the chain nodes is marked with lines of different thicknesses, and the thicker the line is, the larger the transmission strength value is.

[0053] In the preliminary conduction chain structure, straight lines of different thicknesses are used to mark the conduction strength between chain nodes. For example, thicker straight lines are used to connect chain nodes with large conduction strength values, and thinner straight lines are used to connect chain nodes with small conduction strength values, so as to intuitively show the size of the conduction strength.

[0054] Step S129: Perform closed loop detection on the conduction chain structure after marking the conduction strength, delete the conduction path with closed loop, and keep the path with one-way conduction.

[0055] The closed loop detection is performed on the conduction chain structure after marking the conduction strength, that is, it is checked whether there is a circulating conduction path (such as the conduction path A→B→C→A). If there is a conduction path with closed loop, the conduction path is deleted, and only the path with one-way conduction (such as the conduction path A→B→C) is kept, so as to ensure the rationality of the conduction chain.

[0056] Step S1210: Obtain the historical settlement cause conduction record of the slope, compare the currently constructed conduction chain structure with the historical settlement cause conduction record, keep the conduction path consistent with the historical record before the settlement occurs, adjust the conduction strength marking of the conduction path obviously different from the historical record, thereby constructing a complete slope settlement cause conduction chain, the arrangement order of each chain node in the slope settlement cause conduction chain is consistent with the conduction order between the causes, and the conduction strength marking combines the influence of the rock-soil layer type and the historical settlement data.

[0057] The historical settlement cause conduction record of the mountainous road slope is obtained, and the currently constructed conduction chain structure is compared with the historical record. For example, before the settlement occurs in the historical record, the conduction path is surface action cause information→underground action cause information→hydrological action cause information, and the conduction strength marking is a certain value. If the same conduction path exists in the currently constructed conduction chain structure, the conduction path is kept; if the conduction path is obviously different from the historical record (such as the conduction path is surface action cause information→biological action cause information→underground action cause information), the conduction strength marking of the conduction path is adjusted to make it more consistent with the influence of the historical settlement data. In this way, a complete slope settlement cause conduction chain is constructed, the arrangement order of each chain node in the slope settlement cause conduction chain is consistent with the conduction order between the causes, and the conduction strength marking combines the influence of the rock-soil layer type and the historical settlement data.

[0058] Step S130: Generate a cause conduction tracking sequence according to the slope settlement cause conduction chain, the cause conduction tracking sequence records the conduction order between the chain nodes and the state change information in the conduction process.

[0059] Step S131: Extract the unique identification information of all chain nodes in the slope settlement cause conduction chain, which is used to distinguish different cause information, and each unique identification information corresponds to a set of basic attributes including cause type, action area and initial state data.

[0060] In this embodiment, the unique identification information of all chain nodes in the slope settlement cause conduction chain is extracted. For example, the unique identification information of the surface action cause information can be set as "surface-001", and the corresponding basic attributes include the cause type (surface action cause), the action area (slope top area) and the initial state data (such as initial displacement data, initial stress data, etc.); the unique identification information of the underground action cause information can be set as "underground-002", and the corresponding basic attributes include the cause type (underground action cause), the action area (sand layer area) and the initial state data (such as initial stress data), etc.

[0061] Step S132: Arrange the unique identification information of each chain node in turn according to the conduction order of the chain nodes in the slope settlement cause conduction chain to form a basic tracking sequence, and the position of each unique identification information in the basic tracking sequence corresponds to its order in the conduction order.

[0062] According to the conduction order of the chain nodes in the slope settlement cause conduction chain, the unique identification information of each chain node is arranged in turn. For example, the conduction order is surface action cause information ("surface-001") → underground action cause information ("underground-002") → hydrological action cause information ("hydrology-003") → biological action cause information ("biology-004"), and the basic tracking sequence is ["surface-001", "underground-002", "hydrology-003", "biology-004"], and the position of each unique identification information corresponds to its order in the conduction order.

[0063] Step S133: Obtain real-time environmental data of the slope area, and the real-time environmental data includes atmospheric humidity data, ground temperature data and wind force data.

[0064] Real-time environmental data is collected by deploying environmental monitoring equipment such as a weather station in the slope area. The atmospheric humidity sensor of the weather station collects atmospheric humidity data, the temperature sensor collects ground temperature data, and the wind speed and direction sensor collects wind force data (including wind speed and direction), which together constitute the real-time environmental data.

[0065] Step S134: According to the real-time environmental data, filter the state change information collection period of each chain node, shorten the collection interval of state change information when the atmospheric humidity exceeds the normal range, and maintain the normal collection interval when the wind force data is in the stable range.

[0066] According to the real-time environmental data, the state change information collection period of each chain node is screened. For example, when the atmospheric humidity exceeds the normal range (such as atmospheric humidity greater than 80% or less than 20%), it indicates that the water content of the slope soil body may change greatly, and has a greater impact on the slope settlement, so the state change information collection interval is shortened (such as from once every hour to once every half hour); when the wind force data is in a stable range (such as wind speed between 0-5 m / s and wind direction stable), the normal collection interval (such as once every hour) is maintained.

[0067] Step S135: Obtain the state change information of each chain node in the conduction process, which includes the initial state data before conduction, the intermediate state data during conduction, and the final state data after conduction. Each state data is attached with real-time environmental data at the collection time.

[0068] The state change information of each chain node in the conduction process is obtained. Taking the surface action trigger information (“surface-001”) as an example, the initial state data before conduction is the initial displacement data (such as the initial position data collected by the GNSS receiver), the intermediate state data during conduction is the displacement data collected during the conduction process (such as displacement data collected every half hour), and the final state data after conduction is the displacement data at the end of the conduction. Each state data is attached with real-time environmental data at the collection time (such as atmospheric humidity, ground temperature, wind force data at the collection time).

[0069] Step S136: Associate the state change information of each chain node with the corresponding unique identification information in the basic tracking sequence to form a tracking unit with state annotation. Each tracking unit with state annotation includes unique identification information, corresponding state change information, and associated real-time environmental data.

[0070] The state change information of each chain node is associated with the corresponding unique identification information in the basic tracking sequence. For example, the unique identification information of the surface action trigger information is “surface-001”, the corresponding state change information (initial displacement data, intermediate displacement data, final displacement data) and the associated real-time environmental data (atmospheric humidity, ground temperature, wind force data at the collection time) form a tracking unit with state annotation, such as {“surface-001”, [initial displacement data, intermediate displacement data, final displacement data], [atmospheric humidity data, ground temperature data, wind force data]}.

[0071] Step S137: According to the arrangement order of the basic tracking sequence, all tracking units with state annotation are sequentially spliced to form an initial trigger conduction tracking sequence.

[0072] According to the arrangement order of the basic tracking sequence, all the tracking units with state annotations are spliced in turn. For example, the basic tracking sequence is ["surface-001", "underground-002", "hydrology-003", "biology-004"], and the corresponding tracking units with state annotations are { "surface-001",...}, { "underground-002",...}, { "hydrology-003",...}, and { "biology-004",...} in turn. Splicing them in turn forms the initial cause conduction tracking sequence.

[0073] Step S138: Set a tracking update period, the length of which is adjusted according to the change frequency of real-time environmental data. If the change frequency of real-time environmental data is high, the tracking update period is shortened, and if the change frequency of real-time environmental data is low, the tracking update period is lengthened.

[0074] The tracking update period is set. For example, when the change frequency of real-time environmental data (such as atmospheric humidity, ground temperature, and wind data) is high (such as changing once every hour), it indicates that the environmental conditions of the slope change rapidly, and have a greater impact on the slope settlement, so the tracking update period is shortened (such as from once every day to once every half day); when the change frequency of real-time environmental data is low (such as changing once every day), the tracking update period is lengthened (such as from once every half day to once every day).

[0075] Step S139: In each tracking update period, the latest state change information of each chain node is obtained, and real-time environmental data at the update time is collected.

[0076] In each tracking update period, the latest state change information of each chain node is obtained. For example, in each tracking update period of half a day, the latest state data (such as the latest displacement data and stress data) of each chain node such as surface action cause information and underground action cause information is collected, and real-time environmental data (such as atmospheric humidity, ground temperature, and wind data at the update time) is collected.

[0077] Step S1310: The latest state change information of each chain node and the corresponding real-time environmental data are used to replace the old state change information and the old real-time environmental data of the corresponding tracking unit in the initial cause conduction tracking sequence, to generate an updated cause conduction tracking sequence. The cause conduction tracking sequences before and after the update are compared, and the chain node identifiers, state change data differences, and real-time environmental data differences corresponding to the differences are recorded.

[0078] The latest state change information and corresponding real-time environmental data of each chain node are used to replace the old state change information and old real-time environmental data of the corresponding tracking unit in the initial inducement conduction tracking sequence. For example, in the tracking unit of the surface layer action inducement information, the old displacement data is replaced with the latest displacement data, the old atmospheric humidity data is replaced with the latest atmospheric humidity data, and the like, to generate an updated inducement conduction tracking sequence. Then, the inducement conduction tracking sequences before and after the update are compared for differences, and the chain node identification (such as “surface layer-001”), the state change data difference (such as the change amount of displacement data), and the real-time environmental data difference (such as the change amount of atmospheric humidity) corresponding to the difference part are recorded.

[0079] Step S1311: The difference record result is attached to the end of the updated inducement conduction tracking sequence to form a complete inducement conduction tracking sequence, and the complete inducement conduction tracking sequence is stored in layers, and the storage levels are divided according to the tracking update period. Each level stores the update sequence and the difference record in the corresponding period, facilitating subsequent tracing of the conduction state changes in different time periods.

[0080] The difference record result is attached to the end of the updated inducement conduction tracking sequence to form a complete inducement conduction tracking sequence. Then, the complete inducement conduction tracking sequence is stored in layers, and the storage levels are divided according to the tracking update period. For example, each half-day tracking update period corresponds to a storage level, and the update sequence and the difference record in the period are stored to facilitate subsequent tracing of the conduction state changes in different time periods.

[0081] Step S140: Determine the key conduction node of the slope settlement according to the inducement conduction tracking sequence, and the key conduction node is the chain node with the highest degree of influence on the settlement trend.

[0082] Step S141: Extract the state change information of all chain nodes in the inducement conduction tracking sequence, and count the number of state changes of each chain node in the tracking period, and the tracking period is the total length of the inducement conduction tracking sequence record.

[0083] In this embodiment, the state change information of all chain nodes in the inducement conduction tracking sequence is extracted. Taking the surface layer action inducement information (“surface layer-001”) as an example, its state change information includes the displacement data changes before, during and after conduction. The number of state changes of each chain node in the tracking period is counted, and the tracking period is the total length of the inducement conduction tracking sequence record (such as one month). For example, the number of state changes of the surface layer action inducement information in one month is N1, the number of state changes of the underground action inducement information is N2, and the like.

[0084] Step S142: Extract all change amplitude features from the state change information of each chain node, adjust the statistical method of the change amplitude features according to the corresponding cause type, and classify and statistically analyze the change amplitude features of the underground rock and soil stress change information according to the stress direction, and classify and statistically analyze the change amplitude features of the surrounding water body infiltration information according to the infiltration depth.

[0085] All change amplitude features are extracted from the state change information of each chain node. For underground rock and soil stress change information (such as "underground-002"), the change amplitude feature is the difference between the maximum and minimum values of stress change, and the change amplitude feature is classified and statistically analyzed according to the stress direction (such as horizontal stress, vertical stress) combined with the cause type (underground action cause); for surrounding water body infiltration information (such as "hydrology-003"), the change amplitude feature is the difference between the maximum and minimum values of water level or infiltration pressure change, and the change amplitude feature is classified and statistically analyzed according to the infiltration depth (such as shallow infiltration, deep infiltration).

[0086] Step S143: Calculate the state change amplitude mean value of each chain node, add all the classified and statistically analyzed change amplitude features of the chain node, and then divide by the number of state changes to obtain the state change amplitude mean value.

[0087] The state change amplitude mean value of each chain node is calculated. Taking the underground rock and soil stress change information ("underground-002") as an example, the sum of the horizontal stress change amplitude features is S1, the sum of the vertical stress change amplitude features is S2, and the number of state changes is N2. Then the state change amplitude mean value is (S1+S2) / N2.

[0088] Step S144: Obtain the settlement sensitivity partition data of the slope area, which divides the slope into an area where the rock and soil layer compressive strength is less than a preset strength and settlement has occurred in a historical setting period, an area where the rock and soil layer compressive strength is between two preset strengths and no settlement has occurred in a historical setting period but there is a settlement risk, and an area where the rock and soil layer compressive strength is greater than a preset strength and has no historical settlement record.

[0089] The settlement sensitivity partition data of the slope area is obtained through field investigation and laboratory rock and soil mechanical property testing. Field investigation includes drilling core sampling, in-situ testing (such as standard penetration test, static cone penetration test), etc., and laboratory rock and soil mechanical property testing includes compressive strength test, shear strength test, etc. According to the test results, the slope is divided into three types of areas: an area where the rock and soil layer compressive strength is less than a preset strength (such as 20 MPa) and settlement has occurred in a historical setting period (such as the past five years); an area where the rock and soil layer compressive strength is between two preset strengths (such as 20 MPa-50 MPa) and no settlement has occurred in a historical setting period but there is a settlement risk; and an area where the rock and soil layer compressive strength is greater than a preset strength (such as 50 MPa) and has no historical settlement record.

[0090] Step S145: Determine the sensitivity weight according to the settlement sensitivity partition to which the action area of each chain node belongs. The sensitivity weight of the area where the compressive strength of the rock-soil layer is less than the preset strength and settlement has occurred in the historical setting period is greater than that of the area where the compressive strength of the rock-soil layer is between two preset strengths and no settlement has occurred in the historical setting period but there is a settlement risk, and the sensitivity weight of the area where the compressive strength of the rock-soil layer is greater than the preset strength and has no historical settlement record is greater.

[0091] Determine the sensitivity weight according to the settlement sensitivity partition to which the action area of each chain node belongs. For example, the sensitivity weight of the area where the compressive strength of the rock-soil layer is less than 20 MPa and settlement has occurred in the historical five years is set to 3; the sensitivity weight of the area where the compressive strength of the rock-soil layer is between 20 MPa and 50 MPa and no settlement has occurred in the historical five years but there is a settlement risk is set to 2; and the sensitivity weight of the area where the compressive strength of the rock-soil layer is greater than 50 MPa and has no historical settlement record is set to 1.

[0092] Step S146: Calculate the state activity index of each chain node.

[0093] Step S1461: Obtain the state change frequency of each chain node, which is the total number of state changes of the chain node in the inducement conduction tracking sequence recording period, and the determination standard of state change is that the change amplitude feature exceeds the conventional fluctuation range of the inducement type.

[0094] Obtain the state change frequency of each chain node. Taking the surface action inducement information (“surface-001”) as an example, the state change frequency thereof in the inducement conduction tracking sequence recording period (such as one month) is N1, and the determination standard of state change is that the displacement change amplitude exceeds the conventional fluctuation range of the inducement type (surface action inducement) (such as the displacement change amplitude being greater than 5 mm).

[0095] Step S1462: Extract all change amplitude features from the state change information of each chain node, and according to the inducement type corresponding to the chain node, formulate classification and statistical rules, the change amplitude features of the slope surface soil disturbance information are classified according to the disturbance range, the change amplitude features of the underground rock-soil layer stress change information are classified according to the stress value change, the change amplitude features of the surrounding water body permeation information are classified according to the permeation rate change, and the change amplitude features of the vegetation coverage change information are classified according to the coverage area change.

[0096] All change amplitude features are extracted from the state change information of each chain node, and classification statistical rules are formulated according to the corresponding cause type of the chain node. For example, the change amplitude features of the slope surface soil disturbance information are classified according to the disturbance range size (such as small range disturbance, medium range disturbance, and large range disturbance); the change amplitude features of the underground rock and soil layer stress change information are classified according to the stress value change (such as stress increase and stress decrease); the change amplitude features of the surrounding water body permeation information are classified according to the permeation rate change (such as permeation rate increase and permeation rate decrease); and the change amplitude features of the vegetation coverage change information are classified according to the coverage area change (such as coverage area increase and coverage area decrease).

[0097] Step S1463: Classify the change amplitude features of each chain node according to the classification statistical rules, and count the numerical sum and the corresponding change times of each type of change amplitude features.

[0098] The change amplitude features of each chain node are classified according to the classification statistical rules. Taking the underground rock and soil layer stress change information (“underground-002”) as an example, the change amplitude features of stress value change to increase are classified into one category, and the numerical sum of this type of change amplitude features is S21, and the corresponding change times are N21; the change amplitude features of stress value change to decrease are classified into one category, and the numerical sum of this type of change amplitude features is S22, and the corresponding change times are N22.

[0099] Step S1464: Calculate the state change amplitude mean value of each chain node, first calculate the mean value of each type of change amplitude feature, and then determine the classification weight according to the importance of each type of change in the influence of the cause. The classification weight corresponding to the category with high importance degree is greater than the classification weight corresponding to the category with low importance degree.

[0100] The state change amplitude mean value of each chain node is calculated. Taking the underground rock and soil layer stress change information (“underground-002”) as an example, the mean value of the stress increase type change amplitude feature is S21 / N21, and the mean value of the stress decrease type change amplitude feature is S22 / N22. According to the importance of each type of change in the influence of the cause, the classification weight is determined. Since the stress increase may have a greater impact on the slope settlement, the classification weight of the stress increase category is set to 2, and the classification weight of the stress decrease category is set to 1. Then the state change amplitude mean value is (S21 / N21) x 2 + (S22 / N22) x 1.

[0101] Step S1465: Superimpose the coordinate range of the action area of each chain node with the coordinate range of the settlement sensitivity partition data, and output the settlement sensitivity partition to which the action area of the chain node belongs.

[0102] The coordinate range of each chain node action area is superimposed with the coordinate range of the settlement sensitivity partition data, and the settlement sensitivity partition to which the chain node action area belongs is output by the GIS software. For example, after the action area of the surface layer action inducement information (slope top area) is superimposed with the settlement sensitivity partition data, it is output that the settlement sensitivity partition to which it belongs is an area with rock-soil layer compressive strength less than 20 MPa and settlement occurring in the past five years.

[0103] Step S1466: Determine the sensitivity weight according to the settlement sensitivity partition. The sensitivity weight of an area with rock-soil layer compressive strength less than a preset strength and settlement occurring in a historical setting period is a first weight value, the sensitivity weight of an area with rock-soil layer compressive strength between two preset strengths and no settlement occurring in the historical setting period but with settlement hidden danger is a second weight value, and the sensitivity weight of an area with rock-soil layer compressive strength greater than the preset strength and no historical settlement record is a third weight value. The first weight value is greater than the second weight value, and the second weight value is greater than the third weight value.

[0104] The sensitivity weight is determined according to the settlement sensitivity partition. For example, the sensitivity weight (first weight value) of an area with rock-soil layer compressive strength less than 20 MPa and settlement occurring in the past five years is 3; the sensitivity weight (second weight value) of an area with rock-soil layer compressive strength between 20 MPa and 50 MPa and no settlement occurring in the past five years but with settlement hidden danger is 2; and the sensitivity weight (third weight value) of an area with rock-soil layer compressive strength greater than 50 MPa and no historical settlement record is 1.

[0105] Step S1467: If the action area of the chain node spans multiple settlement sensitivity partitions, the area proportion of each partition in the action area is calculated, the sensitivity weight corresponding to each partition is multiplied by the area proportion, and then the sum is obtained to obtain the comprehensive sensitivity weight of the chain node.

[0106] If the action area of the chain node spans multiple settlement sensitivity partitions, the area proportion of each partition in the action area is calculated. For example, the action area of the surface layer action inducement information spans an area with rock-soil layer compressive strength less than 20 MPa and settlement occurring in the past five years (area proportion is 60%) and an area with rock-soil layer compressive strength between 20 MPa and 50 MPa and no settlement occurring in the past five years but with settlement hidden danger (area proportion is 40%), and the comprehensive sensitivity weight is 3*60%+2*40%=2.6.

[0107] Step S1468: Obtain the historical reference change amplitude of the inducement type corresponding to the chain node, which is extracted from the slope historical settlement data and is the average value of the change amplitude of the inducement type in the historical period.

[0108] The historical reference change range of the corresponding cause type of the chain node is obtained. For example, the historical reference change range of the surface action cause information corresponding to the cause type (surface action cause) is M1 (which is the average value of the change range of the cause type in the past five years).

[0109] Step S1469: Calculate the ratio of the state change range average value to the historical reference change range to obtain the standardized change range.

[0110] The ratio of the state change range average value to the historical reference change range is calculated to obtain the standardized change range. Taking the surface action cause information as an example, the state change range average value is M1_current, the historical reference change range is M1, and the standardized change range is M1_current / M1.

[0111] Step S14610: Set the calculation method of the state active index, and the state active index is equal to the product of the state change frequency, the standardized change range, and the sensitivity weight.

[0112] The calculation method of the state active index is set, and the state active index = state change frequency x standardized change range x sensitivity weight.

[0113] Step S14611: Substitute the state change frequency, the state change range average value, and the sensitivity weight of each chain node into the calculation method to obtain the state active index of the chain node.

[0114] The calculation method of the state active index is set, and the state active index = state change frequency x standardized change range x sensitivity weight.

[0115] Step S14612: Record the state active index calculation process data of each chain node, including the state change frequency statistical basis, the state change range average value classification weight, the sensitivity weight value basis, and the final calculation result, to form a state active index calculation file.

[0116] Record the state active index calculation process data of each chain node, including the state change frequency statistical basis (such as the number of times the displacement change range exceeds the normal fluctuation range), the state change range average value classification weight (such as the classification weight of the stress increase type is 2, and the classification weight of the stress reduction type is 1), the sensitivity weight value basis (such as the area ratio of the action region across the partition), and the final calculation result, to form a state active index calculation file for subsequent verification and analysis.

[0117] Step S147: Extract the conduction intensity data of each chain node in the slope settlement cause conduction chain, and calculate the total conduction intensity of each chain node as a conduction initiation cause information. The total conduction intensity is the sum of the conduction intensities between the chain node and all conduction reception cause information. Meanwhile, adjust the total conduction intensity in combination with the settlement sensitivity partition to which the conduction reception cause information belongs. When the action area of the conduction reception cause information is an area where the rock-soil layer compressive strength is less than the preset strength and settlement has occurred in the historical setting period, the statistical weight of the corresponding conduction intensity is greater than that when the action area of the conduction reception cause information is other settlement sensitivity partitions.

[0118] Extract the conduction intensity data of each chain node in the slope settlement cause conduction chain. Take the surface layer action cause information (“surface-001”) as an example, and calculate the total conduction intensity of the surface layer action cause information as a conduction initiation cause information. The total conduction intensity is the sum of the conduction intensities between the chain node and all conduction reception cause information (such as underground action cause information, hydrological action cause information, etc.). Meanwhile, adjust the total conduction intensity in combination with the settlement sensitivity partition to which the conduction reception cause information belongs. For example, when the action area of the conduction reception cause information is an area where the rock-soil layer compressive strength is less than 20 MPa and settlement has occurred in the historical five years, the statistical weight of the corresponding conduction intensity is set to 2; when the action area is other settlement sensitivity partitions, the statistical weight is set to 1.

[0119] Step S148: Calculate the conduction influence index of each chain node. The conduction influence index is the ratio of the total conduction intensity to the state active index.

[0120] Calculate the conduction influence index of each chain node. The conduction influence index = total conduction intensity / state active index. Take the surface layer action cause information as an example. The total conduction intensity is T1, and the state active index is A1. Therefore, the conduction influence index is T1 / A1.

[0121] Step S149: Arrange all chain nodes in descending order of conduction influence index to form a conduction influence index sorting table. The conduction influence index sorting table includes chain node identification, conduction influence index, state active index, and total conduction intensity.

[0122] Arrange all chain nodes in descending order of conduction influence index to form a conduction influence index sorting table. For example, the top few positions of the sorting table are the surface layer action cause information (“surface-001”, conduction influence index I1), underground action cause information (“underground-002”, conduction influence index I2), etc. The sorting table includes chain node identification, conduction influence index, state active index, and total conduction intensity.

[0123] Step S1410: Select the part of the chain nodes arranged in the front as the candidate key conduction nodes of the slope settlement, the number of candidate key conduction nodes is determined according to the total number of chain nodes and the number of settlement sensitivity partitions, the number of candidate key conduction nodes corresponding to the region where the compressive strength of the rock-soil layer is less than the preset strength and settlement occurs in the historical setting period is more than the region where the compressive strength of the rock-soil layer is between two preset strengths and no settlement occurs in the historical setting period but there is a settlement risk and the region where the compressive strength of the rock-soil layer is greater than the preset strength and has no historical settlement record.

[0124] The part of the chain nodes arranged in the front is selected as the candidate key conduction nodes. For example, the total number of chain nodes is 10, there are 3 settlement sensitivity partitions, the number of candidate key conduction nodes corresponding to the region where the compressive strength of the rock-soil layer is less than 20 MPa and settlement occurs in the historical five years is 5, the number of candidate key conduction nodes corresponding to the region where the compressive strength of the rock-soil layer is between 20 MPa-50 MPa and no settlement occurs in the historical five years but there is a settlement risk is 3, and the number of candidate key conduction nodes corresponding to the region where the compressive strength of the rock-soil layer is greater than 50 MPa and has no historical settlement record is 2.

[0125] Step S1411: Obtain the historical settlement data of the slope, which includes the historical settlement occurrence period, the historical settlement amount data, and the state data of each inducement information in the corresponding period.

[0126] The historical settlement data of the mountainous highway slope is obtained, including the historical settlement occurrence period (such as May 2020-June 2020), the historical settlement amount data (such as the settlement amount in the period is 100 mm), and the state data of each inducement information in the corresponding period (such as the state data of the surface action inducement information, the underground action inducement information, etc. in May 2020-June 2020).

[0127] Step S1412: Divide the historical settlement data into different historical settlement scenes according to the settlement amount, and the reference priority of the historical settlement scene where the settlement amount exceeds the first preset value is higher than that of the historical settlement scene where the settlement amount does not exceed the first preset value.

[0128] The historical settlement data is divided into different historical settlement scenes according to the settlement amount. For example, the first preset value is set to 50 mm, the reference priority of the historical settlement scene where the settlement amount exceeds 50 mm (such as the scene where the settlement amount is 100 mm, 80 mm) is higher than that of the historical settlement scene where the settlement amount does not exceed 50 mm (such as the scene where the settlement amount is 30 mm, 40 mm).

[0129] Step S1413: Extracting the cause conduction node data before the settlement occurs from the historical settlement scene, determining the key conduction node identification and the corresponding state change characteristics before the historical settlement occurs, comparing the state change information of the candidate key conduction node in the historical settlement occurrence period with the state change information in the current period, retaining the candidate key conduction node with consistent state change trend, and checking its update frequency in the cause conduction tracking sequence at the same time. The candidate key conduction node with an update frequency exceeding the second preset value is retained in priority to the candidate key conduction node with an update frequency not exceeding the second preset value, thereby determining the key conduction node of the slope settlement, forming a key conduction node list, and the key conduction node list contains node identification, action area, settlement sensitivity partition, state active index and conduction influence index.

[0130] Extracting the cause conduction node data before the settlement occurs from the historical settlement scene. For example, from the historical settlement scene with a settlement amount of 100 mm, the cause conduction node data in May 2020 (one month before the settlement occurs) is extracted, the key conduction node identification (such as "surface layer-001", "underground-002") before the historical settlement occurs and the corresponding state change characteristics (such as the increase of displacement change amplitude of surface layer action cause information, the increase of stress change amplitude of underground action cause information) are determined. The state change information of the candidate key conduction node in the historical settlement occurrence period (such as May 2020-June 2020) is compared with the state change information in the current period (such as May 2024-June 2024). First, the change law of the state change frequency is compared, such as the state change frequency of the surface layer action cause information in the historical period is 10 times, and the state change frequency in the current period is 12 times, whether the change law is consistent is judged; then the increase and decrease law of the state change amplitude average value is compared, such as the state change amplitude average value of the surface layer action cause information in the historical period is 5 mm, and the state change amplitude average value in the current period is 6 mm, whether the increase and decrease law is consistent is judged; finally, the period distribution law of the state change occurrence is compared, such as the state change of the surface layer action cause information mainly occurs in the late period in the historical period, whether it also mainly occurs in the late period in the current period.

[0131] Step S1413-1: Extracting the historical settlement occurrence period from the slope historical settlement data, and determining the start time and end time of the historical settlement occurrence period.

[0132] Extracting the historical settlement occurrence period from the slope historical settlement data, for example, the historical settlement occurrence period is from May 1, 2020 to June 1, 2020, the start time is May 1, 2020 00:00, and the end time is June 1, 2020 00:00.

[0133] Step S1413-2: Extract the state change information of the candidate key conduction node in the historical subsidence occurrence period, including the number of state changes, the starting time of each state change, and the state change amplitude characteristics in the historical subsidence occurrence period.

[0134] The state change information of the candidate key conduction node in the historical subsidence occurrence period is extracted. Taking the candidate key conduction node "surface-001" as an example, during the period from May 1, 2020 to June 1, 2020, the number of state changes is 10, the starting time of each state change is May 5, 2020 00:00, May 10, 2020 00:00, etc., and the state change amplitude characteristics are displacement change amplitudes of 3mm, 5mm, etc.

[0135] Step S1413-3: Arrange the state change information of the candidate key conduction node in the historical subsidence occurrence period in chronological order to form a historical state change sequence.

[0136] The state change information of the candidate key conduction node in the historical subsidence occurrence period is arranged in chronological order to form a historical state change sequence. For example, the historical state change sequence of "surface-001" is [(2020-05-05 00:00, 3mm), (2020-05-10 00:00, 5mm),...].

[0137] Step S1413-4: Determine the starting time and ending time of the current period, and the duration of the current period is consistent with the duration of the historical subsidence occurrence period.

[0138] The starting time and ending time of the current period are determined, and the duration of the current period is consistent with the duration of the historical subsidence occurrence period. For example, the current period is from May 1, 2024 to June 1, 2024, the starting time is May 1, 2024 00:00, the ending time is June 1, 2024 00:00, and the duration is one month, which is consistent with the duration of the historical subsidence occurrence period.

[0139] Step S1413-5: Extract the state change information of the candidate key conduction node in the current period, including the number of state changes, the starting time of each state change, and the state change amplitude characteristics.

[0140] The state change information of the candidate key conduction node in the current period is extracted. Taking "surface-001" as an example, during the period from May 1, 2024 to June 1, 2024, the number of state changes is 12, the starting time of each state change is May 6, 2024 00:00, May 11, 2024 00:00, etc., and the state change amplitude characteristics are displacement change amplitudes of 4mm, 6mm, etc.

[0141] Step S1413-6: Arrange the state change information of the candidate key conduction node in the current period in chronological order to form a current state change sequence.

[0142] Arrange the state change information of the candidate key conduction node in the current period in chronological order to form a current state change sequence. For example, the current state change sequence of “Surface-001” is [ (2024-05-06 00:00, 4mm), (2024-05-11 00:00, 6mm),... ].

[0143] Step S1413-7: Calculate the change trend similarity between the historical state change sequence and the current state change sequence. The change trend similarity is obtained by comparing the change law of the number of state changes, the increase and decrease law of the state change amplitude, and the change law of the state change interval length in the historical state change sequence and the current state change sequence.

[0144] Calculate the change trend similarity between the historical state change sequence and the current state change sequence. Compare the change law of the number of state changes: the number of state changes in the historical period is 10 times, and the number of state changes in the current period is 12 times, the change law is consistent (both show a growing trend); compare the increase and decrease law of the state change amplitude: the state change amplitude in the historical period is 3mm, 5mm (increasing), and the state change amplitude in the current period is 4mm, 6mm (increasing), the increase and decrease law is consistent; compare the change law of the state change interval length: the average state change interval length in the historical period is 5 days, and the average state change interval length in the current period is 4.5 days, the change law is relatively consistent. Based on the above factors, calculate the change trend similarity (e.g. similarity is 0.9).

[0145] Step S1413-8: When the change trend similarity is greater than or equal to the similarity threshold, it is determined that the state change trend of the candidate key conduction node is consistent, the candidate key conduction node with consistent state change trend is retained, and the candidate key conduction node with change trend similarity less than the similarity threshold is eliminated.

[0146] Set the similarity threshold to 0.8. Since the change trend similarity is 0.9, which is greater than 0.8, it is determined that the state change trend of the candidate key conduction node (“Surface-001”) is consistent, and the candidate key conduction node is retained; if the change trend similarity is less than 0.8, the candidate key conduction node is eliminated.

[0147] Step S1413-9: For the retained candidate key conduction node, recheck its conduction influence index to ensure that the conduction influence index is still in the front row of the conduction influence index ranking table, and finally determine the key conduction node of the slope settlement.

[0148] The retained candidate key conduction nodes are checked again for their conduction influence indexes. For example, the conduction influence index of "Surface Layer-001" is still in the top 3 in the conduction influence index ranking table, and thus it is finally determined as a key conduction node of the slope settlement.

[0149] The retained candidate key conduction nodes are checked again for their conduction influence indexes to ensure that the conduction influence indexes are still in the top (e.g., top 5) in the conduction influence index ranking table, and are checked for their update frequencies in the cause conduction tracking sequence. The second preset value is set to be updated once every half day, and the candidate key conduction nodes with an update frequency exceeding the second preset value (e.g., updated once every 3 hours) are retained in priority to the candidate key conduction nodes with an update frequency not exceeding the second preset value. Thus, the key conduction nodes of the slope settlement are determined, a key conduction node list is formed, and the list contains node identifiers (e.g., "Surface Layer-001", "Underground-002"), action areas (e.g., slope top area, sand layer area), settlement sensitivity subareas (e.g., an area with rock and soil layer compressive strength less than 20 MPa and settlement occurring in the past five years), state active indexes, and conduction influence indexes.

[0150] Step S150: Based on the positions and state change information of the key conduction nodes, a monitoring resource directional deployment scheme is generated, which contains monitoring device deployment positions and monitoring frequency settings for the key conduction nodes.

[0151] Step S151: The cause information types corresponding to the key conduction nodes are extracted, and the monitoring device types required for each cause information type are determined. The monitoring device types match the cause information types and are used to collect corresponding types of cause information.

[0152] In this embodiment, the cause information types corresponding to the key conduction nodes are extracted. For example, the cause information type corresponding to the key conduction node "Surface Layer-001" is surface layer action cause information (slope surface soil body disturbance information), and the required monitoring device type is a GNSS receiver, which is used to collect displacement data of the slope surface layer; the cause information type corresponding to the key conduction node "Underground-002" is underground action cause information (underground rock and soil layer stress change information), and the required monitoring device type is a vibrating wire stress sensor, which is used to collect stress data of the underground rock and soil layer.

[0153] Step S152: The position information of the key conduction nodes is obtained, which contains longitude and latitude coordinates and specific area identifiers of the slopes.

[0154] The position information of the key conduction nodes is obtained. For example, the position information of the key conduction node "surface layer-001" is the latitude and longitude coordinates (X1, Y1), and the specific area of the slope where the key conduction node is located is identified as the slope top area A; the position information of the key conduction node "underground-002" is the latitude and longitude coordinates (X2, Y2), and the specific area of the slope where the key conduction node is located is identified as the sand layer area B.

[0155] Step S153: According to the position information of the key conduction nodes, the monitoring area is divided, and each monitoring area corresponds to one or more key conduction nodes.

[0156] According to the position information of the key conduction nodes, the monitoring area is divided. For example, the slope top area A (containing the key conduction node "surface layer-001") is divided into a monitoring area 1, and the sand layer area B (containing the key conduction node "underground-002") is divided into a monitoring area 2; or the slope top area A and the adjacent slope area (containing other key conduction nodes) are divided into a monitoring area, and each monitoring area corresponds to one or more key conduction nodes.

[0157] Step S154: In each monitoring area, the terrain features are analyzed, including the slope, the slope direction and the surface flatness.

[0158] In each monitoring area, the terrain features are analyzed. Taking the monitoring area 1 (the slope top area A) as an example, through the digital elevation model (DEM) data or field measurement, the slope (such as the slope is 15°), the slope direction (such as the slope direction is south) and the surface flatness (such as the surface flatness is good, and the fluctuation is less than 5 cm) of the area are obtained.

[0159] Step S155: According to the terrain features, the deployment point of the monitoring equipment is selected, and the deployment point needs to meet the conditions that the monitoring equipment is not blocked and can effectively collect the trigger information of the corresponding key conduction node.

[0160] According to the terrain features, the deployment point of the monitoring equipment is selected. For the monitoring area 1 (the slope top area A), since the slope is 15°, the slope direction is south, and the surface flatness is good, a GNSS receiver is selected to be deployed in the open area on the slope top, so as to ensure that the receiver is not blocked (such as there are no tall buildings or trees around) and can effectively collect the displacement data of the slope surface layer; for the monitoring area 2 (the sand layer area B), a buried hole of a vibrating wire stress sensor is selected to be deployed on the surface above the sand layer, so as to ensure that the sensor can effectively collect the stress data of the underground sand layer.

[0161] Step S156: The number of monitoring equipment at each deployment point is determined, and the number of monitoring equipment is determined according to the state active index of the key conduction node, and the number of monitoring equipment deployed is more when the state active index value is larger.

[0162] Determine the number of monitoring devices for each deployment site. Take the key conduction node "surface-001" as an example. If the state active index A1 is large (e.g., A1>10), deploy 2 GNSS receivers at its deployment site; if A1 is small (e.g., A1<5), deploy 1 GNSS receiver.

[0163] Step S157: Extract the state change information of the key conduction node and count the state change frequency of the key conduction node in a unit time.

[0164] Extract the state change information of the key conduction node and count the state change frequency of the key conduction node in a unit time. For example, the state change frequency of the key conduction node "surface-001" in a unit time (e.g., one day) is F1 (e.g., 3 times of state change in one day).

[0165] Step S158: Set the monitoring frequency of the monitoring device according to the state change frequency, assign a unique device identifier to the monitoring device in each monitoring area, and record the association relationship between the device identifier and the corresponding key conduction node.

[0166] Set the monitoring frequency of the monitoring device according to the state change frequency. For example, the state change frequency of the key conduction node "surface-001" is F1 (3 times of state change in one day), and the monitoring frequency of the GNSS receiver is set to collect data every 8 hours; assign a unique device identifier (e.g., GNSS-001, GNSS-002) to the GNSS receiver in the monitoring area 1, and record the association relationship between the device identifier and the corresponding key conduction node "surface-001".

[0167] Step S159: Integrate the monitoring area division result, deployment site coordinates, monitoring device type and number, monitoring frequency setting, and device identifier association relationship to generate a monitoring resource directional deployment scheme.

[0168] The monitoring area division result (such as monitoring area 1: slope top area A, monitoring area 2: sand layer area B), deployment point coordinates (such as the deployment point coordinates of GNSS-001 are (X1, Y1), and the deployment point coordinates of the vibrating wire stress sensor are (X2, Y2)), the monitoring device type and quantity (such as 2 GNSS receivers are deployed in monitoring area 1, and 1 vibrating wire stress sensor is deployed in monitoring area 2), the monitoring frequency setting (such as the GNSS receiver collects data every 8 hours, and the vibrating wire stress sensor collects data every 12 hours), and the device identification association (such as GNSS-001 is associated with "surface-001", and vibrating wire stress sensor-001 is associated with "underground-002") are integrated to generate a monitoring resource directional deployment scheme, which will serve as the basis for subsequent slope settlement monitoring resource deployment, ensuring that monitoring resources are accurately directed to key conduction nodes and improving the efficiency and accuracy of slope settlement monitoring.

[0169] Step S1510: According to the position information of the key conduction nodes, the monitoring area is divided, including:

[0170] Step S1510-1: Import the position information of all key conduction nodes into a geographic information system, and generate a position marker point for each key conduction node in the geographic information system.

[0171] In this embodiment, the position information (latitude and longitude coordinates) of the key conduction nodes "surface-001", "underground-002", etc. is imported into a geographic information system (such as ArcGIS software), and a position marker point for each key conduction node is generated in the software. The style of the marker point can be set to different colors or shapes to distinguish different key conduction nodes.

[0172] Step S1510-2: Analyze the spatial distance between each position marker point and calculate the straight-line distance between any two position marker points.

[0173] The spatial distance between each position marker point is analyzed, and the straight-line distance between any two position marker points is calculated. For example, the position marker point coordinates of the key conduction node "surface-001" are (X1, Y1), and the position marker point coordinates of the key conduction node "underground-002" are (X2, Y2). The straight-line distance D between them is calculated according to the distance formula between two points: D = √[(X2-X1)²+(Y2-Y1)²].

[0174] Step S1510-3: When the straight-line distance between two position marker points is less than or equal to the distance threshold, the two position marker points are classified into the same group.

[0175] A distance threshold (e.g. 50m) is set. If the straight-line distance between two position marker points is less than or equal to 50m, the two position marker points are grouped into the same group. For example, the straight-line distance between the position marker points of the key conduction nodes "Surface-001" and "Surface-003" is 40m, which is less than the distance threshold 50m. Therefore, they are grouped into the same group.

[0176] Step S1510-4: Repeat the grouping operation described above to group all position marker points according to spatial distance relationships, forming a plurality of position marker point groups. For each position marker point group, calculate the coordinate mean of all position marker points in the position marker point group to obtain the center coordinate of the position marker point group.

[0177] Repeat the grouping operation of step S1510-3 to group all position marker points according to spatial distance relationships, forming a plurality of position marker point groups. For each position marker point group, calculate the coordinate mean of all position marker points in the group to obtain the center coordinate of the position marker point group. For example, the position marker point group contains "Surface-001" (X1, Y1) and "Surface-003" (X3, Y3). The center coordinate is ((X1+X3) / 2, (Y1+Y3) / 2).

[0178] Step S1510-5: Draw a circular region with the center coordinate of each position marker point group as the center and the distance threshold as the radius. If there is an overlapping part between adjacent circular regions, combine the corresponding position marker point groups and recalculate the center coordinate and the circular region of the combined group.

[0179] Draw a circular region with the center coordinate of each position marker point group as the center and the distance threshold (50m) as the radius. If there is an overlapping part between adjacent circular regions (e.g. the overlapping area of two circular regions is greater than 0), combine the corresponding position marker point groups and recalculate the center coordinate and the circular region of the combined group. For example, there is an overlap between the circular region of position marker point group 1 and the circular region of position marker point group 2. Combine the two groups and recalculate the center coordinate of the combined group ((X1+X2+X3+X4) / 4, (Y1+Y2+Y3+Y4) / 4). Then draw a new circular region with the center coordinate as the center and the distance threshold as the radius.

[0180] Step S1510-6: Repeat the merging operation until there is no overlapping part between all circular regions. Determine each non-overlapping circular region as a monitoring region. Each monitoring region corresponds to a group of position marker points, i.e. one or more key conduction nodes.

[0181] The merging operation is repeated until there is no overlapping part in all the circular regions. Each non-overlapping circular region is determined as a monitoring region, and each monitoring region corresponds to a group of position marker points, i.e., corresponds to one or more key conduction nodes. For example, monitoring region 1 (containing key conduction nodes "surface-001", "surface-003") and monitoring region 2 (containing key conduction node "underground-002") are finally formed.

[0182] Step S1510-7: Record the boundary coordinates, center coordinates and corresponding key conduction node identifiers of each monitoring region, associate the monitoring region division result with the administrative division information of the slope, supplement the administrative division identifier where the monitoring region is located, and perfect the monitoring region division data.

[0183] Record the boundary coordinates (the boundary coordinates of the circular region can be calculated by the center coordinates and the radius), the center coordinates and the corresponding key conduction node identifiers (such as monitoring region 1 corresponding to "surface-001", "surface-003") of each monitoring region. Associate the monitoring region division result with the administrative division information of the slope, supplement the administrative division identifier (such as a certain county and a certain town) where the monitoring region is located, and perfect the monitoring region division data for subsequent management and statistics.

[0184] Step S158 includes:

[0185] For example, step S158-1: Extract the state change frequency of the key conduction node in a unit time, and the state change frequency is the number of state changes of the key conduction node in a preset unit time.

[0186] Extract the state change frequency of the key conduction node in a unit time. Taking the key conduction node "surface-001" as an example, the preset unit time is one day, and the number of state changes in one day is 3 times, so the state change frequency is 3 times / day.

[0187] Step S158-2: Divide the state change frequency into multiple intervals, and each interval corresponds to a monitoring frequency level.

[0188] Divide the state change frequency into multiple intervals. For example, the state change frequency 0-1 times / day corresponds to the monitoring frequency level 1, 1-3 times / day corresponds to the monitoring frequency level 2, and 3-5 times / day corresponds to the monitoring frequency level 3.

[0189] Step S158-3: Set a first frequency interval, and when the state change frequency of the key conduction node is in the first frequency interval, it corresponds to a first monitoring frequency level.

[0190] Set the first frequency interval to 3-5 times / day, and when the state change frequency of the key conduction node is in this interval, it corresponds to the first monitoring frequency level (monitoring frequency level 3).

[0191] Step S158-4: Set a second frequency interval, the upper limit value of the second frequency interval is equal to the lower limit value of the first frequency interval, when the state change frequency of the key conducting node is in the second frequency interval, it corresponds to a second monitoring frequency level, the monitoring frequency of the second monitoring frequency level is lower than the first monitoring frequency level.

[0192] The second frequency interval is set to 1-3 times / day, the upper limit value (3 times / day) is equal to the lower limit value (3 times / day) of the first frequency interval, when the state change frequency of the key conducting node is in the interval, it corresponds to a second monitoring frequency level (monitoring frequency level 2), the monitoring frequency of the monitoring frequency level 2 is lower than the monitoring frequency level 3 (for example, the monitoring time interval of the monitoring frequency level 3 is 4 hours, and the monitoring time interval of the monitoring frequency level 2 is 8 hours).

[0193] Step S158-5: Set multiple frequency intervals and corresponding monitoring frequency levels in the above manner, and the smaller the value of the frequency interval, the lower the corresponding monitoring frequency level.

[0194] Set multiple frequency intervals and corresponding monitoring frequency levels in the above manner, such as 0-1 times / day corresponding to monitoring frequency level 1 (monitoring time interval is 12 hours), and the smaller the value of the frequency interval, the lower the corresponding monitoring frequency level.

[0195] Step S158-6: Substitute the state change frequency of each key conducting node into the frequency interval division standard to determine the corresponding monitoring frequency level of the key conducting node.

[0196] Substitute the state change frequency of each key conducting node into the frequency interval division standard. For example, the state change frequency of the key conducting node "surface-001" is 3 times / day, which is in the interval of 1-3 times / day, corresponding to monitoring frequency level 2; the state change frequency of the key conducting node "underground-002" is 2 times / day, which is also in the interval of 1-3 times / day, corresponding to monitoring frequency level 2.

[0197] Step S158-7: Determine the specific monitoring time interval according to the monitoring frequency level, the monitoring time interval is the reciprocal of the monitoring frequency, and the higher the monitoring frequency level, the shorter the monitoring time interval.

[0198] Determine the specific monitoring time interval according to the monitoring frequency level. The monitoring frequency of the monitoring frequency level 2 is 1 / 8 times / hour (i.e. monitoring once every 8 hours), and the monitoring time interval is 8 hours; the monitoring frequency of the monitoring frequency level 3 is 1 / 4 times / hour (i.e. monitoring once every 4 hours), and the monitoring time interval is 4 hours.

[0199] Step S158-8: Record the corresponding monitoring frequency level and monitoring time interval of each key conducting node.

[0200] The monitoring frequency level and monitoring time interval corresponding to each key conduction node are recorded. For example, the monitoring frequency level of "Surface-001" is 2, and the monitoring time interval is 8 hours; the monitoring frequency level of "Underground-002" is 2, and the monitoring time interval is 8 hours.

[0201] Step S158-9: If a monitoring area contains multiple key conduction nodes, the highest monitoring frequency level corresponding to the key conduction nodes is taken as the monitoring frequency level of the monitoring area.

[0202] If a monitoring area contains multiple key conduction nodes, the highest monitoring frequency level corresponding to the key conduction nodes is taken as the monitoring frequency level of the monitoring area. For example, monitoring area 1 contains "Surface-001" (monitoring frequency level 2) and "Surface-003" (monitoring frequency level 3), and the monitoring frequency level of the monitoring area is 3, and the monitoring time interval is 4 hours.

[0203] Step S158-10: According to the monitoring frequency level of the monitoring area, the monitoring frequency of all monitoring devices in the monitoring area is determined, and the sampling interval of the monitoring devices is adjusted to match the state change frequency of the key conduction nodes.

[0204] According to the monitoring frequency level of the monitoring area, the monitoring frequency of all monitoring devices in the monitoring area is determined. The monitoring frequency level of monitoring area 1 is 3, and the monitoring time interval is 4 hours, so the sampling interval of the monitoring devices such as GNSS receivers in this area is adjusted to 4 hours to match the state change frequency of the key conduction nodes, ensuring that the state change information of the key conduction nodes can be captured in time.

[0205] Through the above steps, the slope settlement monitoring method provided by the embodiment starts from the acquisition of multi-source cause information, goes through the construction of cause conduction chain, the generation of cause conduction tracking sequence, the determination of key conduction nodes, and finally generates a monitoring resource directional deployment scheme, realizes the accurate monitoring of slope settlement and the reasonable deployment of resources, and can effectively improve the efficiency and accuracy of slope settlement monitoring. In practical application, the parameters (such as sampling frequency, distance threshold, similarity threshold, etc.) in each step can be adjusted according to the specific slope situation and monitoring demand to optimize the monitoring effect.

[0206] Based on the same inventive concept, please refer to Figure 2 , which shows the structure schematic block diagram of the slope settlement monitoring system 100 provided by the embodiment of the present application for executing the above-mentioned slope settlement monitoring method. The slope settlement monitoring system 100 can include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0207] In this embodiment, the machine readable storage medium 120 can also be integrated into the processor 130, and can communicate and interact with external systems through the communication unit 110. The machine readable storage medium 120 is configured to store machine executable instructions for implementing the solutions of the present application, and the processor 130 is configured to execute the machine executable instructions stored in the machine readable storage medium 120 to implement the slope settlement monitoring method provided by the foregoing method embodiments.

[0208] It should be noted that, in order to simplify the description of the present application and to facilitate the understanding of one or more embodiments of the present application, in the foregoing description of the embodiments of the present application, various features are sometimes combined into one embodiment, figure or description thereof.

Claims

1. A method of monitoring settlement of a slope, characterized by, The method comprises: obtaining multi-source inducing information of a slope area, the multi-source inducing information including surface soil disturbance information, underground rock and soil stress change information, surrounding water body penetration information and vegetation coverage change information; constructing a slope subsidence inducing conduction chain based on the multi-source inducing information, the slope subsidence inducing conduction chain taking each inducing information as a chain node, and the connection relationship between the chain nodes embodying the subsidence influence conduction relationship between the inducements; generating an inducing conduction tracking sequence according to the slope subsidence inducing conduction chain, the inducing conduction tracking sequence recording the conduction sequence between the chain nodes and the state change information in the conduction process; determining a key conduction node of slope subsidence according to the inducing conduction tracking sequence, the key conduction node being the chain node with the highest influence degree on the subsidence trend; generating a monitoring resource directional deployment scheme based on the position and state change information of the key conduction node, the monitoring resource directional deployment scheme including the monitoring equipment deployment position and monitoring frequency setting for the key conduction node; the constructing of the slope subsidence inducing conduction chain based on the multi-source inducing information comprises: classifying the multi-source inducing information according to the action level of influencing the slope subsidence to obtain surface action inducing information, underground action inducing information, hydrological action inducing information and biological action inducing information, the surface action inducing information corresponding to the surface soil disturbance information, the underground action inducing information corresponding to the underground rock and soil stress change information, the hydrological action inducing information corresponding to the surrounding water body penetration information, and the biological action inducing information corresponding to the vegetation coverage change information; extracting state change features in each type of inducing information, the state change features including a change starting time, a change duration and a change amplitude feature; obtaining rock and soil layer type distribution data of the slope area, the rock and soil layer type distribution data including a clay layer distribution range, a sandy soil layer distribution range, a weathered rock layer distribution range and an intact rock layer distribution range; adjusting the extraction precision of the state change features according to the rock and soil layer type distribution data, improving the extraction precision of the change amplitude feature in the clay layer distribution range, and maintaining the basic extraction precision in the intact rock layer distribution range; comparing the state change features of different categories of inducing information, when the change starting time of one type of inducing information is earlier than that of another type of inducing information, and there is an overlapping period in the change duration of the two types of inducing information, combining the rock and soil layer type distribution data to determine whether there is a spatial intersection in the action area of the two types of inducing information; if there is a spatial intersection, marking the former as a conduction initiation inducing information and the latter as a conduction reception inducing information; if there is no spatial intersection, rechecking the overlapping period of the change duration of the two types of inducing information, and determining again after extending the overlapping period determination period; taking the conduction initiation inducing information and the conduction reception inducing information as chain nodes respectively, connecting the two chain nodes with a straight line to embody the conduction relationship, and forming a preliminary conduction chain structure; The connection frequency between the same transmission initiation cause information and the transmission reception cause information in the preliminary transmission chain structure is counted statistically, and the statistical weight of the connection frequency is adjusted according to the influence degree of the rock-soil layer type on the transmission, so that the statistical weight of the connection frequency in the clay layer distribution area is greater than that in the complete rock layer distribution area; The transmission strength between the chain nodes is determined according to the connection frequency after the adjustment, and the connection frequency value is greater, the transmission strength value is greater; In the preliminary transmission chain structure, the transmission strength between the chain nodes is marked by straight lines of different thicknesses, and the thicker the straight line, the greater the transmission strength value; The transmission chain structure after the transmission strength is marked is subjected to closed loop detection, the transmission path with closed loop is deleted, and the path with one-way transmission is reserved; The historical settlement cause transmission record of the slope is obtained, the transmission chain structure currently constructed is compared with the historical settlement cause transmission record, the transmission path consistent with the settlement before the occurrence in the historical record is reserved, the transmission strength marking of the transmission path obviously different from the historical record is adjusted, and thus a complete slope settlement cause transmission chain is constructed, the arrangement order of each chain node in the slope settlement cause transmission chain is consistent with the transmission order between the causes, and the transmission strength marking is combined with the influence of the rock-soil layer type and the historical settlement data; The generation of the cause transmission tracking sequence according to the slope settlement cause transmission chain comprises the following steps: Extracting unique identification information of all chain nodes in the slope settlement cause transmission chain, the unique identification information being used for distinguishing different cause information, each unique identification information corresponding to a group of basic attributes including cause type, action area and initial state data; Arranging the unique identification information of each chain node in turn according to the transmission order of the chain nodes in the slope settlement cause transmission chain to form a basic tracking sequence, and the position of each unique identification information in the basic tracking sequence corresponding to the order thereof in the transmission order; Obtaining real-time environment data of the slope area, the real-time environment data including atmospheric humidity data, ground temperature data and wind force data; Filtering the state change information collection period of each chain node according to the real-time environment data, shortening the collection interval of the state change information when the atmospheric humidity exceeds the normal range, and maintaining the normal collection interval when the wind force data is in the stable range; Obtaining state change information of each chain node in the transmission process, the state change information including initial state data before transmission, intermediate state data in transmission and final state data after transmission, and each state data being attached with real-time environment data at the collection time; Associating the state change information of each chain node with the corresponding unique identification information in the basic tracking sequence to form a tracking unit with state marking, each tracking unit with state marking including unique identification information, corresponding state change information and associated real-time environment data; Arranging all the tracking units with state marking in turn according to the arrangement order of the basic tracking sequence to form an initial cause transmission tracking sequence; Setting a tracking update period, the length of the tracking update period being adjusted according to the change frequency of the real-time environment data, the tracking update period being shortened when the change frequency of the real-time environment data is high, and the tracking update period being lengthened when the change frequency of the real-time environment data is low. In each tracking update cycle, the latest state change information of each chain node is obtained, and real-time environmental data at the update time is collected; The latest state change information and the corresponding real-time environmental data of each chain node are used to replace the old state change information and the old real-time environmental data of the corresponding tracking unit in the initial inducement conduction tracking sequence, to generate an updated inducement conduction tracking sequence. The inducement conduction tracking sequences before and after the update are compared, and the chain node identifier, state change data difference and real-time environmental data difference corresponding to the difference are recorded; The difference record result is attached to the end of the updated inducement conduction tracking sequence to form a complete inducement conduction tracking sequence. The complete inducement conduction tracking sequence is stored in layers. The storage levels are divided according to the tracking update cycle. Each level stores the update sequence and the difference record in the corresponding period, which facilitates the subsequent tracing of the conduction state change in different periods. The state change characteristics of different categories of inducement information are compared. When the change start time of one category of inducement information is earlier than that of another category of inducement information, and there is an overlapping period in the change duration of the two categories of inducement information, the spatial intersection of the action areas of the two categories of inducement information is determined in combination with the rock-soil layer type distribution data, including: The change start time and the change duration of each category of inducement information are extracted from the state change characteristics of each category of inducement information. The change end time of each category of inducement information is calculated. The change end time is the sum of the change start time and the change duration. The action area coordinate range of each category of inducement information is extracted. The action area coordinate range is determined according to the collection points and the influence range of the multi-source inducement information. The action area coordinate range of the surface action inducement information corresponds to the surrounding area of the collection point of the surface soil disturbance. The action area coordinate range of the underground action inducement information corresponds to the surrounding area of the monitoring hole of the underground rock-soil layer stress change. The action area coordinate range of each category of inducement information is superimposed with the rock-soil layer type distribution data to determine the rock-soil layer types contained in the action area of each category of inducement information and the proportion of each type of rock-soil layer. Any two categories of inducement information are selected and marked as the first category of inducement information and the second category of inducement information. The change start time of the first category of inducement information is compared with the change start time of the second category of inducement information. When the change start time of the first category of inducement information is earlier than that of the second category of inducement information, the time difference between the change end time of the first category of inducement information and the change start time of the second category of inducement information is calculated. If the time difference is less than zero, it is determined that the change end time of the first category of inducement information is later than the change start time of the second category of inducement information, and there is an overlapping period in the change duration of the two categories of inducement information. At this time, the action area coordinate range of the first category of inducement information and the action area coordinate range of the second category of inducement information are extracted, and the intersection area of the two coordinate ranges is calculated. The ratio of the intersection area to the total area of the action area of the first category of inducement information is taken as the spatial intersection proportion. At the same time, the rock-soil layer types corresponding to the intersection area are checked. If the proportion of clay or sandy soil in the intersection area exceeds the preset proportion, it is determined that there is an effective spatial intersection between the action areas of the two categories of inducement information. If the spatial intersection ratio is greater than zero and there is an effective spatial intersection, the first type of inducement information is marked as a conduction initiation inducement information, and the second type of inducement information is marked as a conduction reception inducement information; If the spatial intersection ratio is zero or the effective rock-soil layer ratio in the intersection area does not reach the preset ratio, the overlap period of the change duration of the two types of inducement information is re-verified, the time difference between the change end time and the change start time is calculated again after the overlap period is extended, and the overlap situation is re-determined; When the change start time of the first type of inducement information is later than the change start time of the second type of inducement information, the steps of calculating the time difference, the spatial intersection ratio, and the effective rock-soil layer ratio are repeated, and if there is an overlap period and the action area has an effective spatial intersection, the second type of inducement information is marked as a conduction initiation inducement information, and the first type of inducement information is marked as a conduction reception inducement information; The above operations are repeated for any combination of two types of inducement information, and the rock-soil layer type data and spatial intersection data in each judgment process are recorded to form a conduction relationship judgment record, which facilitates subsequent verification of the rationality of the conduction relationship; After completing the conduction initiation and reception marking of all types of inducement information, the marking results are summarized to form a preliminary conduction initiation and reception relationship table, which includes conduction initiation inducement information identification, conduction reception inducement information identification, overlap period data, and spatial intersection data.

2. The method of claim 1, wherein, The key conduction nodes of the slope settlement are determined according to the inducement conduction tracking sequence, which includes: Extracting the state change information of all chain nodes in the inducement conduction tracking sequence, and counting the number of state changes of each chain node in the tracking period, which is the total duration recorded by the inducement conduction tracking sequence; Extracting all change amplitude features from the state change information of each chain node, adjusting the change amplitude feature statistics according to the inducement type corresponding to the chain node, and classifying and counting the change amplitude features of the underground rock-soil layer stress change information according to the stress direction, and classifying and counting the change amplitude features of the surrounding water permeation information according to the permeation depth; Calculating the average state change amplitude of each chain node, adding all the classified and counted change amplitude features of this chain node, and then dividing by the number of state changes to obtain the average state change amplitude; Obtaining the settlement sensitivity partition data of the slope area, which divides the slope into an area with rock-soil layer compressive strength less than a preset strength and settlement occurring in a historical setting period, an area with rock-soil layer compressive strength between two preset strengths and no settlement occurring in a historical setting period but with settlement risks, and an area with rock-soil layer compressive strength greater than a preset strength and no historical settlement record; determining a sensitivity weight according to a settlement sensitivity partition to which an action area of each chain node belongs, the sensitivity weight of a region where the compressive strength of the geotechnical layer is less than a preset strength and settlement has occurred in a historical setting period being greater than that of a region where the compressive strength of the geotechnical layer is between two preset strengths and no settlement has occurred in the historical setting period but there is a settlement risk, the sensitivity weight of the region where the compressive strength of the geotechnical layer is between the two preset strengths and no settlement has occurred in the historical setting period but there is a settlement risk being greater than that of a region where the compressive strength of the geotechnical layer is greater than the preset strength and there is no historical settlement record; calculating a state active index of each chain node; extracting the conduction intensity data of each chain node in the slope settlement inducing conduction chain, and calculating the total conduction intensity of each chain node as a conduction initiating inducing information, the total conduction intensity being the sum of the conduction intensities between the chain node and all conduction receiving inducing information, and the total conduction intensity being adjusted in combination with the settlement sensitivity partition to which the conduction receiving inducing information belongs, the statistical weight of the corresponding conduction intensity being greater when the action area of the conduction receiving inducing information is a region where the compressive strength of the geotechnical layer is less than a preset strength and settlement has occurred in a historical setting period than when the action area of the conduction receiving inducing information is a region of other settlement sensitivity partitions; calculating a conduction influence index of each chain node, the conduction influence index being the ratio of the total conduction intensity to the state active index; arranging all chain nodes in descending order of the conduction influence index to form a conduction influence index sorting table, the conduction influence index sorting table including chain node identification, conduction influence index, state active index and total conduction intensity; selecting part of the chain nodes arranged in the front as candidate key conduction nodes of the slope settlement, the number of the candidate key conduction nodes being determined according to the total number of the chain nodes and the number of the settlement sensitivity partitions, the number of the candidate key conduction nodes corresponding to a region where the compressive strength of the geotechnical layer is less than a preset strength and settlement has occurred in a historical setting period being greater than that of a region where the compressive strength of the geotechnical layer is between two preset strengths and no settlement has occurred in the historical setting period but there is a settlement risk and that of a region where the compressive strength of the geotechnical layer is greater than the preset strength and there is no historical settlement record; obtaining slope historical settlement data, the slope historical settlement data including a historical settlement occurring period, historical settlement amount data and state data of each inducing information in the corresponding period; dividing the historical settlement data into different historical settlement scenarios according to the settlement amount, the reference priority of a historical settlement scenario where the settlement amount exceeds a first preset value being higher than that of a historical settlement scenario where the settlement amount does not exceed the first preset value; The cause conduction node data before the settlement occurs is extracted from the historical settlement scene, the key conduction node identification before the settlement occurs and the corresponding state change characteristics are determined, the state change information of the candidate key conduction node in the historical settlement occurrence period is compared with the state change information in the current period, the candidate key conduction node with consistent state change trend is retained, and the update frequency of the candidate key conduction node in the cause conduction tracking sequence is checked, the candidate key conduction node with the update frequency exceeding the second preset value is retained in priority to the candidate key conduction node with the update frequency not exceeding the second preset value, thereby the key conduction node of the slope settlement is determined, the key conduction node list is formed, and the key conduction node list includes the node identification, the action area, the settlement sensitivity partition, the state active index and the conduction influence index.

3. The method of claim 2, wherein, The state active index of each chain node is calculated, including: The state change times of each chain node are obtained, the state change times are the total number of state changes of the chain node in the cause conduction tracking sequence recording period, and the determination standard of the state change is that the change amplitude characteristic exceeds the conventional fluctuation range of the cause type; All change amplitude characteristics are extracted from the state change information of each chain node, classification statistical rules are formulated according to the cause type corresponding to the chain node, the change amplitude characteristics of the slope surface soil disturbance information are classified according to the disturbance range, the change amplitude characteristics of the underground rock and soil stress change information are classified according to the stress value change, the change amplitude characteristics of the surrounding water body permeation information are classified according to the permeation rate change, and the change amplitude characteristics of the vegetation coverage change information are classified according to the coverage area change; The change amplitude characteristics of each chain node are classified according to the classification statistical rules, and the numerical sum and the corresponding change times of each type of change amplitude characteristic are counted; The state change amplitude mean of each chain node is calculated, the mean of each type of change amplitude characteristic is calculated first, and then the classification weight is determined according to the importance of each type of change in the cause influence, and the classification weight corresponding to the classification with high importance is greater than the classification weight corresponding to the classification with low importance; The mean of each type of change amplitude characteristic is multiplied by the corresponding classification weight, and then summed, to obtain the comprehensive change amplitude mean of the chain node, that is, the state change amplitude mean; The settlement sensitivity partition data of the slope area are obtained, the settlement sensitivity partition data are obtained through field investigation and laboratory rock and soil layer mechanical property test, the determination standard of the area with rock and soil layer compressive strength less than the preset strength and settlement occurring in the historical setting period is that the rock and soil layer compressive strength is less than the preset strength and settlement occurs in the historical setting period, the determination standard of the area with rock and soil layer compressive strength between two preset strengths and no settlement occurring in the historical setting period but with settlement hidden danger is that the rock and soil layer compressive strength is between two preset strengths and no settlement occurs in the historical setting period but with settlement hidden danger, and the determination standard of the area with rock and soil layer compressive strength greater than the preset strength and no historical settlement record is that the rock and soil layer compressive strength is greater than the preset strength and no historical settlement record; The coordinate range of the action area of each chain node is superimposed with the coordinate range of the settlement sensitivity partition data, and the settlement sensitivity partition to which the action area of the chain node belongs is output. According to the sensitivity partitioning of the settlement, the sensitivity weight is determined, the sensitivity weight of the area where the compressive strength of the rock-soil layer is less than the preset strength and settlement occurs in the historical setting period is a first weight value, the sensitivity weight of the area where the compressive strength of the rock-soil layer is between two preset strengths and no settlement occurs in the historical setting period but there is a settlement hidden danger is a second weight value, and the sensitivity weight of the area where the compressive strength of the rock-soil layer is greater than the preset strength and has no historical settlement record is a third weight value, the first weight value is greater than the second weight value, and the second weight value is greater than the third weight value; If the action area of the chain node crosses multiple settlement sensitivity partitions, the area proportion of each partition in the action area is calculated, the sensitivity weight of each partition is multiplied by the area proportion, and then the sum is obtained, to obtain the comprehensive sensitivity weight of the chain node; The historical reference change amplitude of the cause type corresponding to the chain node is obtained, which is extracted from the historical settlement data of the slope and is the average value of the change amplitude of the cause type in the historical period; The ratio of the average change amplitude to the historical reference change amplitude is calculated to obtain the standardized change amplitude; The calculation method of the state activity index is set, and the state activity index is equal to the product of the state change frequency, the standardized change amplitude and the sensitivity weight; The state change frequency, the average change amplitude and the sensitivity weight of each chain node are substituted into the calculation method to obtain the state activity index of the chain node; The state activity index calculation process data of each chain node is recorded, including the state change frequency statistical basis, the average change amplitude classification weight, the sensitivity weight value basis and the final calculation result, to form the state activity index calculation file.

4. The method of claim 1, wherein, Based on the position and state change information of the key conduction node, a monitoring resource directional deployment scheme is generated, including: The cause information type corresponding to the key conduction node is extracted, and the required monitoring device type of each cause information type is determined, the monitoring device type matches the cause information type and is used to collect the corresponding type of cause information; The position information of the key conduction node is obtained, and the position information includes the longitude and latitude coordinates and the specific area identifier of the slope; According to the position information of the key conduction node, the monitoring area is divided, and each monitoring area corresponds to one or more key conduction nodes; In each monitoring area, the terrain features are analyzed, and the terrain features include the slope, the slope direction and the ground flatness; According to the terrain features, the deployment points of the monitoring devices are selected, and the deployment points need to meet the conditions that the monitoring devices are not blocked and can effectively collect the cause information of the corresponding key conduction node; The number of monitoring devices at each deployment point is determined, and the number of monitoring devices is determined according to the state activity index of the key conduction node, and the more the number of deployed monitoring devices is, the larger the state activity index value is; The state change information of the key conduction node is extracted, and the state change frequency of the key conduction node in a unit time is counted; According to the state change frequency, the monitoring frequency of the monitoring device is set, a unique device identifier is allocated to the monitoring device in each monitoring area, and the association relationship between the device identifier and the corresponding key conduction node is recorded. The monitoring area division result, the deployment point coordinate, the monitoring device type and quantity, the monitoring frequency setting and the device identification association relationship are integrated to generate a monitoring resource directional deployment scheme.

5. The method of claim 4, wherein, The monitoring area is divided according to the position information of the key conduction node, and the monitoring area division result is obtained. The position information of all key conduction nodes is imported into a geographic information system, and a position marker point of each key conduction node is generated in the geographic information system. The spatial distances between the position marker points are analyzed, and the straight line distances between any two position marker points are calculated. When the straight line distance between two position marker points is less than or equal to a distance threshold, the two position marker points are grouped into the same group. The grouping operation is repeated, and all position marker points are grouped according to the spatial distance relationship to form a plurality of position marker point groups. For each position marker point group, the coordinate mean value of all position marker points in the position marker point group is calculated to obtain the center coordinate of the position marker point group. The center coordinate of each position marker point group is taken as the center of a circle, and the distance threshold is taken as the radius to draw a circular area. If there is an overlapping part between adjacent circular areas, the corresponding position marker point groups are combined, and the center coordinates and circular areas of the combined groups are recalculated. The combination operation is repeated until there is no overlapping part in all circular areas. Each non-overlapping circular area is determined as a monitoring area, and each monitoring area corresponds to a group of position marker points, that is, one or more key conduction nodes. The boundary coordinates and center coordinates of each monitoring area and the corresponding key conduction node identifier are recorded. The monitoring area division result is associated with the administrative division information of the slope, the administrative division identifier of the monitoring area is supplemented, and the monitoring area division data is perfected.

6. The method of claim 2, wherein, The state change information of the candidate key conduction node in the historical settlement occurrence period is compared with the state change information in the current period, and the candidate key conduction node with consistent state change trend is retained, and the method comprises the following steps: The historical settlement occurrence period is extracted from the historical settlement data of the slope, and the start time and end time of the historical settlement occurrence period are determined. The state change information of the candidate key conduction node in the historical settlement occurrence period is extracted, including the state change times in the historical settlement occurrence period, the start time of each state change and the state change amplitude feature. The state change information of the candidate key conduction node in the historical settlement occurrence period is arranged in time sequence to form a historical state change sequence. The start time and end time of the current period are determined, and the length of the current period is consistent with the length of the historical settlement occurrence period. The state change information of the candidate key conduction node in the current period is extracted, including the state change times, the start time of each state change and the state change amplitude feature. The state change information of the candidate key conduction node in the current period is arranged in time sequence to form a current state change sequence. The change trend similarity of the historical state change sequence and the current state change sequence is calculated. The change trend similarity is obtained by comparing the change law of the state change times, the increase and decrease law of the state change amplitude and the change law of the state change interval length in the historical state change sequence and the current state change sequence. When the change trend similarity is greater than or equal to the similarity threshold, it is determined that the state change trend of the candidate key conduction node is consistent, the candidate key conduction node with the consistent state change trend is retained, and the candidate key conduction node with the change trend similarity less than the similarity threshold is eliminated; For the retained candidate key conduction node, the conduction influence index thereof is checked again to ensure that the conduction influence index is still in the front row of the conduction influence index ranking table, and the key conduction node of the slope settlement is finally determined.

7. A system for monitoring settlement of a slope, the system comprising: Comprise: A processor; A machine readable storage medium for storing machine executable instructions of the processor; Wherein, the processor is configured to execute the machine executable instructions to perform the slope settlement monitoring method in any one of claims 1 to 6.

Images