Self-adaptive individualized landslide prediction method, device and equipment

Abstract

The invention provides a self-adaptive individualized landslide prediction method, device and equipment, and relates to the technical field of landslide prediction, and the method comprises the steps: obtaining the side slope topographic data of a target side slope; constructing a numerical simulation model according to the slope topographic data, calculating a safety coefficient of a target slope, performing sliding surface screening through the safety coefficient to obtain at least one potential sliding surface, and extracting polyline coordinate point information of each potential sliding surface; constructing a sliding surface calculation model, inputting the multi-section line coordinate point information into the sliding surface calculation model to obtain a calculation model sample, and calculating monitoring point displacement data of the side slope surface corresponding to the potential sliding surface through the calculation model sample; and constructing a neural network prediction model, training the neural network prediction model through the monitoring point displacement data and the multi-section line coordinate point information, and predicting the sliding surface of the target slope through the trained neural network prediction model. The pertinence and accuracy of landslide prediction are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of landslide prediction, and in particular to an adaptive individualized landslide prediction method, device and equipment. Background Art

[0002] Traditional landslide early warning schemes mostly rely on specific thresholds set by experience or industry specifications, which are often determined based on existing specifications or past experience. However, in actual applications, there are significant differences in geological conditions, structural characteristics, and environmental factors between different slopes, which makes it difficult for a single early warning threshold to apply to all situations, resulting in poor versatility and flexibility of the early warning system, and prone to false alarms or missed alarms.

[0003] In recent years, neural network models have shown great potential in the field of landslide early warning. However, the training of neural network models usually requires a large amount of representative historical data, and the differences between different slopes are significant. Even for the same type of slopes, the deformation curves of their monitoring data may be very different. Therefore, the generalization ability of the model is limited, and when faced with new or unseen data, the prediction effect is poor, affecting the accuracy of the early warning. In addition, a single model is difficult to adapt to the diversity of slopes and changes in their internal structure and external environment. The model cannot effectively reflect the true state of the slope, reducing the effectiveness and reliability of the early warning. Summary of the invention

[0004] The purpose of the present invention is to provide an adaptive individualized landslide prediction method, device and equipment to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows: In a first aspect, the present application provides an adaptive individualized landslide prediction method, comprising: Acquire slope topographic data of the target slope, wherein the slope topographic data includes geometric characteristic data, geological characteristic data, strength parameters, fault and joint information, and groundwater level information; Constructing a numerical simulation model according to the slope terrain data, calculating the safety factor of the target slope based on the numerical simulation model, screening the sliding surface by the safety factor to obtain at least one potential sliding surface, and extracting the polyline coordinate point information of each potential sliding surface; Constructing a sliding surface calculation model, inputting the multi-segment line coordinate point information of each potential sliding surface into the sliding surface calculation model, obtaining a calculation model sample corresponding to each potential sliding surface, and calculating the displacement data of the monitoring point on the slope surface corresponding to the potential sliding surface through the calculation model sample; A neural network prediction model is constructed, the neural network prediction model is trained by the displacement data of the monitoring points and the coordinate point information of the multi-segment line, and the sliding surface of the target slope is predicted by the trained neural network prediction model.

[0005] In a second aspect, the present application also provides an adaptive individualized landslide prediction device, comprising: An acquisition module, used to acquire slope terrain data of a target slope, wherein the slope terrain data includes geometric feature data, geological feature data, strength parameters, fault and joint information, and groundwater level information; A simulation module, used for constructing a numerical simulation model according to the slope terrain data, calculating the safety factor of the target slope based on the numerical simulation model, screening the sliding surface by the safety factor to obtain at least one potential sliding surface, and extracting the polyline coordinate point information of each potential sliding surface; A calculation module is used to construct a sliding surface calculation model, input the multi-segment line coordinate point information of each potential sliding surface into the sliding surface calculation model, obtain a calculation model sample corresponding to each potential sliding surface, and calculate the displacement data of the monitoring point on the slope surface corresponding to the potential sliding surface through the calculation model sample; The prediction module is used to construct a neural network prediction model, train the neural network prediction model through the monitoring point displacement data and the multi-segment line coordinate point information, and predict the sliding surface of the target slope through the trained neural network prediction model.

[0006] In a third aspect, the present application also provides an adaptive individualized landslide prediction device, comprising: Memory for storing computer programs; A processor is used to implement the steps of the adaptive individualized landslide prediction method when executing the computer program.

[0007] The beneficial effects of the present invention are as follows: the present invention adopts a one-slope-one-model strategy for actual engineering slopes, establishes an independent neural network prediction model for each slope, ensures that the model can accurately reflect the characteristics of a single slope, and improves the accuracy and pertinence of early warning; according to the geological information of the slope, a large number of calculation model samples of different potential sliding surfaces based on the actual case are constructed by combining numerical simulation with discrete elements, and a large number of slope behavior data under working conditions are simulated, so that the model can maintain good performance under a wide range of conditions, and improve its generalization ability and robustness; a neural network is used to establish an intelligent prediction model that fully conforms to the actual slope, thereby improving the flexibility and accuracy of landslide early warning.

[0008] Other features and advantages of the present invention will be set forth in the following description, and in part will become apparent from the description, or may be understood by practicing embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 It is a schematic diagram of the process flow of the adaptive individualized landslide prediction method described in an embodiment of the present invention; Figure 2 is a schematic diagram of all potential slip surfaces in an embodiment of the present invention; Figure 3 A schematic diagram of a sliding surface calculation model in an embodiment of the present invention; Figure 4 A schematic diagram of a calculation model sample in an embodiment of the present invention; Figure 5 A schematic diagram showing a comparison between an actual sliding surface and a predicted sliding surface in an embodiment of the present invention; Figure 6 It is a schematic diagram of the structure of the adaptive individualized landslide prediction device described in an embodiment of the present invention.

[0011] Markings in the figure: 800, adaptive individualized landslide prediction device; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0013] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0014] Embodiment 1: This embodiment provides an adaptive individualized landslide prediction method.

[0015] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300, and step S400.

[0016] Step S100: acquiring slope topographic data of a target slope, wherein the slope topographic data includes geometric characteristic data, geological characteristic data, strength parameters, fault and joint information, and groundwater level information; Obtain the slope topographic data of the target slope, including: The geometric characteristic data of the target slope are obtained by laser radar scanning and drone photography, wherein the geometric characteristic data includes the shape, slope, slope direction, elevation and stratum thickness information of the target slope; Obtaining geological characteristic data of the target slope through on-site investigation and sampling, wherein the geological characteristic data includes lithology, structure and fracture data of the underground rock and soil body; In this embodiment, the distribution, thickness, inclination and other information of the strata are understood through geological characteristic data, and samples of the slope rock and soil are obtained through drilling, excavation and other methods.

[0017] Determine the strength parameters of the target slope through in-situ testing and geotechnical testing, wherein the strength parameters include density, elastic modulus, Poisson's ratio, cohesion, internal friction angle and tensile strength of the soil; In this embodiment, in-situ tests and geotechnical tests such as direct shear test, triaxial compression test and unconfined compression test are used to determine the strength parameters.

[0018] Obtaining fault and joint information of the target slope by resistivity imaging and seismic reflection method, wherein the fault and joint information includes the position, strike, dip and inclination of the fault and joint; The groundwater level information of the target slope is obtained by pressure and ultrasonic sensors, and the groundwater level information includes water level, pressure and permeability coefficient.

[0019] In order to ensure that the numerical simulation model established later can truly reflect the geological information of the on-site slope, the on-site in-situ test and other tests need to be carried out at least three times to ensure the accuracy of the parameters. In addition, in order to prevent the omission of weak structures, the structural information such as faults and joints of the entire slope needs to be explored.

[0020] Step S200: constructing a numerical simulation model according to the slope terrain data, calculating the safety factor of the target slope based on the numerical simulation model, screening the sliding surface by the safety factor to obtain at least one potential sliding surface, and extracting the polyline coordinate point information of each potential sliding surface; The step S200 specifically includes the following steps: Constructing a numerical simulation model of the target slope based on the geometric characteristic data, geological characteristic data, strength parameters, fault and joint information, and groundwater level information; Based on the numerical simulation model, the safety factor of the target slope is calculated by the Morgenstern method; Screening the sliding surfaces whose safety factors in the numerical simulation model are lower than a preset threshold to obtain at least one potential sliding surface; Extracting geometric parameter information of each potential sliding surface; The geometric parameter information is parsed to extract the polyline coordinate point information of each potential sliding surface.

[0021] In this embodiment, the slope terrain data is drawn into the slope DXF file by AUtoCAD, imported into the Slide software to build a numerical model, and the safety factor is calculated using the Morgenstern method to screen out all potential sliding surfaces with a safety factor of less than 1.5, such as Figure 2 As shown in the figure, all potential sliding surfaces are screened out, among which the darkest one is the potential sliding surface with the smallest safety factor.

[0022] These potential sliding surfaces are considered to be the most unstable parts. The geometric parameter information of all potential sliding surfaces is exported, which contains the key data of all potential sliding surfaces.

[0023] The geometric parameter information is parsed to extract all the polyline coordinate point information that constitute each sliding surface in the key data. These polyline coordinate point information represent the specific location and shape of the potential sliding surface.

[0024] Step S300: constructing a sliding surface calculation model, inputting the multi-segment line coordinate point information of each potential sliding surface into the sliding surface calculation model, obtaining a calculation model sample corresponding to each potential sliding surface, and calculating the displacement data of the monitoring point on the slope surface corresponding to the potential sliding surface through the calculation model sample; The step S300 specifically includes the following steps: Establishing a slope model based on the geometric characteristic data and fault and joint information; Performing grid division through the slope model to obtain slope grid point data; The slope grid point data is modeled by discrete element method, and the strength parameter is input to obtain a sliding surface calculation model; Inputting the polyline coordinate point information of each potential sliding surface into the sliding surface calculation model respectively, to obtain the calculation model sample corresponding to each potential sliding surface; Select monitoring points on the target slope; Based on the calculation model sample, the monitoring point displacement data of the slope surface from the initial state to the instability process is calculated by the strength reduction method, and the monitoring point displacement data includes the cumulative displacement, velocity, acceleration and displacement deflection angle of the monitoring point.

[0025] In this embodiment, the geometric feature data and fault and joint information are used to establish a slope model through Rhino software, and the grid is divided to obtain the slope grid point data. The slope grid point data is modeled through discrete element software 3dec, and the strength parameters are input to obtain the sliding surface calculation model, such as Figure 3 shown.

[0026] Then, the geometric parameter information of different potential sliding surfaces, that is, the coordinate point information of the polyline constituting the potential sliding surface, is input into the sliding surface calculation model to obtain different calculation model samples, such as Figure 4 Shown are sample examples of the calculation model for four different sliding surfaces of the slope.

[0027] Step S400: constructing a neural network prediction model, training the neural network prediction model through the displacement data of the monitoring points and the coordinate point information of the multi-segment line, and predicting the sliding surface of the target slope through the trained neural network prediction model.

[0028] The step S400 specifically includes the following steps: Build a neural network prediction model; The displacement data of monitoring points are used as the input of the neural network prediction model, and the coordinate point information and intensity parameters of the polyline are used as the output of the neural network prediction model; Dividing all the computational model samples into a training set and a test set; Based on the long short-term memory network, the neural network prediction model is trained and tested by the training set and the test set to obtain a trained neural network prediction model; Acquire actual monitoring point displacement data of the monitoring point, wherein the actual monitoring point displacement data includes the cumulative displacement, velocity, acceleration and displacement deflection angle actually monitored at the monitoring point; The displacement data of the actual monitoring points are input into the trained neural network prediction model for prediction, and the polyline coordinate point information and intensity parameters of the predicted sliding surface are obtained.

[0029] In this embodiment, before using the monitoring point displacement data as the input of the neural network prediction model, it also includes: cleaning the monitoring point displacement data to remove invalid or abnormal data points, and standardizing the data to meet the requirements of the neural network prediction model input.

[0030] In this embodiment, all the calculation model samples are divided according to the ratio of 5:1, that is, 83.3% of the samples are used for training and 16.7% of the samples are used for prediction. By dividing the calculation model samples into training sets and test sets, the prediction accuracy of the model can be improved, so that the model has higher prediction ability and better adaptability.

[0031] Based on the long short-term memory network, the neural network prediction model is trained and tested using the training set and the test set to obtain the trained neural network prediction model.

[0032] In this embodiment, a long short-term memory network is used for model training. The long short-term memory network is good at processing time series data. The long short-term memory network is used to capture the time dependence of slope displacement and deformation data, thereby improving the flexibility and accuracy of the early warning system.

[0033] The accuracy of the trained neural network prediction model is evaluated. Evaluation indicators such as cross-validation, confusion matrix, and recall rate are used to comprehensively measure the performance of the model in predicting the geometric parameters and strength parameters of the slope sliding surface. The specific steps include: Cross-validation is used to ensure the generalization ability of the neural network prediction model on different data sets and reduce the risk of overfitting; a confusion matrix is ​​constructed to show the difference between the prediction results of the neural network prediction model and the true label, and the accuracy is calculated to evaluate the proportion of correct predictions of the neural network prediction model as a whole; the recall rate is introduced to evaluate the ability of the neural network prediction model to identify positive examples (i.e. correctly predict the sliding surface parameters), so as to comprehensively and objectively evaluate the accuracy of the trained neural network prediction model.

[0034] like Figure 5 As shown in the figure, an actual slope example is obtained, and the slope topographic data of the slope is obtained. Through the above steps, a neural network prediction model is constructed for the slope. According to the monitoring data prediction screen of the actual example slope monitoring points JC01, JC02, and JC03, it can be seen that the difference between the actual sliding surface and the predicted sliding surface is small, indicating that the method of the present invention is reliable and effective.

[0035] Embodiment 2: This embodiment provides an adaptive individualized landslide prediction device, the device comprising: An acquisition module, used to acquire slope terrain data of a target slope, wherein the slope terrain data includes geometric feature data, geological feature data, strength parameters, fault and joint information, and groundwater level information; A simulation module, used for constructing a numerical simulation model according to the slope terrain data, calculating the safety factor of the target slope based on the numerical simulation model, screening the sliding surface by the safety factor to obtain at least one potential sliding surface, and extracting the polyline coordinate point information of each potential sliding surface; A calculation module is used to construct a sliding surface calculation model, input the multi-segment line coordinate point information of each potential sliding surface into the sliding surface calculation model, obtain a calculation model sample corresponding to each potential sliding surface, and calculate the displacement data of the monitoring point on the slope surface corresponding to the potential sliding surface through the calculation model sample; The prediction module is used to construct a neural network prediction model, train the neural network prediction model through the monitoring point displacement data and the multi-segment line coordinate point information, and predict the sliding surface of the target slope through the trained neural network prediction model.

[0036] The acquisition module comprises: A first acquisition unit is used to acquire geometric feature data of a target slope by laser radar scanning and drone photography, wherein the geometric feature data includes the shape, slope, slope direction, elevation and stratum thickness information of the target slope; A second acquisition unit is used to acquire geological characteristic data of the target slope through on-site investigation and sampling, wherein the geological characteristic data includes lithology, structure and fracture data of the underground rock and soil body; A third acquisition unit is used to determine the strength parameters of the target slope through on-site in-situ testing and geotechnical testing, wherein the strength parameters include density, elastic modulus, Poisson's ratio, cohesion, internal friction angle and tensile strength of the soil; A fourth acquisition unit is used to acquire fault and joint information of the target slope by resistivity imaging and seismic reflection method, wherein the fault and joint information includes the position, strike, dip and inclination of the fault and joint; The fifth acquisition unit is used to acquire the groundwater level information of the target slope through pressure sensors and ultrasonic sensors, wherein the groundwater level information includes water level, pressure and permeability coefficient.

[0037] The simulation module comprises: A first model building unit is used to build a numerical simulation model of the target slope based on the geometric feature data, geological feature data, strength parameters, fault and joint information, and groundwater level information; A first calculation unit is used to calculate the safety factor of the target slope by the Morgenstern method based on the numerical simulation model; A screening unit, used for screening the sliding surfaces whose safety factors in the numerical simulation model are lower than a preset threshold value, to obtain at least one potential sliding surface; a sixth acquisition unit, configured to extract geometric parameter information of each of the potential sliding surfaces; The parsing unit is used to parse the geometric parameter information and extract the polyline coordinate point information of each potential sliding surface.

[0038] The calculation module comprises: A second model building unit is used to build a slope model according to the geometric feature data and the fault and joint information; A grid division unit, used for performing grid division through the slope model to obtain slope grid point data; A third model building unit is used to model the slope grid point data by discrete element method and input the strength parameter to obtain a sliding surface calculation model; A sample acquisition unit, used for inputting the polyline coordinate point information of each potential sliding surface into the sliding surface calculation model to obtain the calculation model sample corresponding to each potential sliding surface; A monitoring point selection unit is used to select monitoring points of the target slope; The second calculation unit is used to calculate the monitoring point displacement data of the slope surface from the initial state to the instability process of the monitoring point by the strength reduction method based on the calculation model sample, and the monitoring point displacement data includes the cumulative displacement, velocity, acceleration and displacement deflection angle of the monitoring point.

[0039] The prediction module comprises: A fourth model building unit, used for building a neural network prediction model; A fifth model building unit is used to use the displacement data of the monitoring point as the input of the neural network prediction model, and use the coordinate point information and the intensity parameter of the polyline as the output of the neural network prediction model; A sample division unit, used for dividing all the computing model samples into a training set and a test set; A training and testing unit, used for training and testing the neural network prediction model based on a long short-term memory network through the training set and the test set to obtain a trained neural network prediction model; A seventh acquisition unit, used to acquire actual monitoring point displacement data of the monitoring point, wherein the actual monitoring point displacement data includes the accumulated displacement, velocity, acceleration and displacement deflection angle actually monitored at the monitoring point; The prediction unit is used to input the displacement data of the actual monitoring points into the trained neural network prediction model for prediction, and obtain the multi-segment line coordinate point information and intensity parameters of the predicted sliding surface.

[0040] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0041] Embodiment 3: Corresponding to the above method embodiment, this embodiment further provides an adaptive individualized landslide prediction device. The adaptive individualized landslide prediction device described below and the adaptive individualized landslide prediction method described above can refer to each other.

[0042] Figure 6 FIG. 8 is a block diagram of an adaptive individualized landslide prediction device 800 according to an exemplary embodiment. Figure 6 As shown, the adaptive individualized landslide prediction device 800 may include: a processor 801 and a memory 802. The adaptive individualized landslide prediction device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0043] The processor 801 is used to control the overall operation of the adaptive personalized landslide prediction device 800 to complete all or part of the steps in the adaptive personalized landslide prediction method. The memory 802 is used to store various types of data to support the operation of the adaptive personalized landslide prediction device 800, and these data may include, for example, instructions for any application or method operating on the adaptive personalized landslide prediction device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or sent via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the adaptive individualized landslide prediction device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.

[0044] In an exemplary embodiment, the adaptive individualized landslide prediction device 800 can be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), digital signal processors (Digital Signal Processor, referred to as DSP), digital signal processing devices (Digital Signal Processing Device, referred to as DSPD), programmable logic devices (Programmable Logic Device, referred to as PLD), field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA), controllers, microcontrollers, microprocessors or other electronic components, and is used to execute the above-mentioned adaptive individualized landslide prediction method.

[0045] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0046] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An adaptive individualized landslide prediction method, characterized in that: include: Acquire slope topographic data of the target slope, wherein the slope topographic data includes geometric characteristic data, geological characteristic data, strength parameters, fault and joint information, and groundwater level information; Constructing a numerical simulation model according to the slope terrain data, calculating the safety factor of the target slope based on the numerical simulation model, screening the sliding surface by the safety factor to obtain at least one potential sliding surface, and extracting the polyline coordinate point information of each potential sliding surface; Constructing a sliding surface calculation model, inputting the multi-segment line coordinate point information of each potential sliding surface into the sliding surface calculation model, obtaining a calculation model sample corresponding to each potential sliding surface, and calculating the displacement data of the monitoring point on the slope surface corresponding to the potential sliding surface through the calculation model sample; A neural network prediction model is constructed, the neural network prediction model is trained by the displacement data of the monitoring points and the coordinate point information of the multi-segment line, and the sliding surface of the target slope is predicted by the trained neural network prediction model.

2. The adaptive individualized landslide prediction method according to claim 1, characterized in that: Obtain the slope topographic data of the target slope, including: The geometric characteristic data of the target slope are obtained by laser radar scanning and drone photography, wherein the geometric characteristic data includes the shape, slope, slope direction, elevation and stratum thickness information of the target slope; Obtaining geological characteristic data of the target slope through on-site investigation and sampling, wherein the geological characteristic data includes lithology, structure and fracture data of the underground rock and soil body; Determine the strength parameters of the target slope through in-situ testing and geotechnical testing, wherein the strength parameters include density, elastic modulus, Poisson's ratio, cohesion, internal friction angle and tensile strength of the soil; Obtaining fault and joint information of the target slope by resistivity imaging and seismic reflection method, wherein the fault and joint information includes the position, strike, dip and inclination of the fault and joint; The groundwater level information of the target slope is obtained by pressure and ultrasonic sensors, and the groundwater level information includes water level, pressure and permeability coefficient.

3. The adaptive individualized landslide prediction method according to claim 1, characterized in that: A numerical simulation model is constructed according to the slope terrain data, a safety factor of the target slope is calculated based on the numerical simulation model, a sliding surface is screened by the safety factor to obtain at least one potential sliding surface, and polyline coordinate point information of each potential sliding surface is extracted, including: Constructing a numerical simulation model of the target slope based on the geometric characteristic data, geological characteristic data, strength parameters, fault and joint information, and groundwater level information; Based on the numerical simulation model, the safety factor of the target slope is calculated by the Morgenstern method; Screening the sliding surfaces whose safety factors in the numerical simulation model are lower than a preset threshold to obtain at least one potential sliding surface; Extracting geometric parameter information of each potential sliding surface; The geometric parameter information is parsed to extract the polyline coordinate point information of each potential sliding surface.

4. The adaptive individualized landslide prediction method according to claim 1, characterized in that: Constructing a sliding surface calculation model, inputting the multi-segment line coordinate point information of each potential sliding surface into the sliding surface calculation model, obtaining a calculation model sample corresponding to each potential sliding surface, and calculating the displacement data of the monitoring point on the slope surface corresponding to the potential sliding surface through the calculation model sample, including: Establishing a slope model based on the geometric characteristic data and fault and joint information; Performing grid division through the slope model to obtain slope grid point data; The slope grid point data is modeled by discrete element method, and the strength parameter is input to obtain a sliding surface calculation model; Inputting the polyline coordinate point information of each potential sliding surface into the sliding surface calculation model respectively, to obtain the calculation model sample corresponding to each potential sliding surface; Select monitoring points on the target slope; Based on the calculation model sample, the monitoring point displacement data of the slope surface from the initial state to the instability process is calculated by the strength reduction method, and the monitoring point displacement data includes the cumulative displacement, velocity, acceleration and displacement deflection angle of the monitoring point.

5. The adaptive individualized landslide prediction method according to claim 1, characterized in that: Constructing a neural network prediction model, and training the neural network prediction model through the monitoring point displacement data and the multi-segment line coordinate point information, including: Build a neural network prediction model; The displacement data of monitoring points are used as the input of the neural network prediction model, and the coordinate point information and intensity parameters of the polyline are used as the output of the neural network prediction model; Dividing all the computational model samples into a training set and a test set; Based on the long short-term memory network, the neural network prediction model is trained and tested through the training set and the test set to obtain a trained neural network prediction model.

6. An adaptive individualized landslide prediction device, characterized in that: include: An acquisition module, used to acquire slope terrain data of a target slope, wherein the slope terrain data includes geometric feature data, geological feature data, strength parameters, fault and joint information, and groundwater level information; A simulation module, used for constructing a numerical simulation model according to the slope terrain data, calculating the safety factor of the target slope based on the numerical simulation model, screening the sliding surface by the safety factor to obtain at least one potential sliding surface, and extracting the polyline coordinate point information of each potential sliding surface; A calculation module is used to construct a sliding surface calculation model, input the multi-segment line coordinate point information of each potential sliding surface into the sliding surface calculation model, obtain a calculation model sample corresponding to each potential sliding surface, and calculate the displacement data of the monitoring point on the slope surface corresponding to the potential sliding surface through the calculation model sample; The prediction module is used to construct a neural network prediction model, train the neural network prediction model through the monitoring point displacement data and the multi-segment line coordinate point information, and predict the sliding surface of the target slope through the trained neural network prediction model.

7. The adaptive individualized landslide prediction device according to claim 6, characterized in that: The acquisition module comprises: A first acquisition unit is used to acquire geometric feature data of a target slope by laser radar scanning and drone photography, wherein the geometric feature data includes the shape, slope, slope direction, elevation and stratum thickness information of the target slope; A second acquisition unit is used to acquire geological characteristic data of the target slope through on-site investigation and sampling, wherein the geological characteristic data includes lithology, structure and fracture data of the underground rock and soil body; A third acquisition unit is used to determine the strength parameters of the target slope through on-site in-situ testing and geotechnical testing, wherein the strength parameters include density, elastic modulus, Poisson's ratio, cohesion, internal friction angle and tensile strength of the soil; A fourth acquisition unit is used to acquire fault and joint information of the target slope by resistivity imaging and seismic reflection method, wherein the fault and joint information includes the position, strike, dip and inclination of the fault and joint; The fifth acquisition unit is used to acquire the groundwater level information of the target slope through pressure sensors and ultrasonic sensors, wherein the groundwater level information includes water level, pressure and permeability coefficient.

8. The adaptive individualized landslide prediction device according to claim 6, characterized in that: The simulation module comprises: A first model building unit is used to build a numerical simulation model of the target slope based on the geometric feature data, geological feature data, strength parameters, fault and joint information, and groundwater level information; A first calculation unit is used to calculate the safety factor of the target slope by the Morgenstern method based on the numerical simulation model; A screening unit, used for screening the sliding surfaces whose safety factors in the numerical simulation model are lower than a preset threshold value, to obtain at least one potential sliding surface; a sixth acquisition unit, configured to extract geometric parameter information of each of the potential sliding surfaces; The parsing unit is used to parse the geometric parameter information and extract the polyline coordinate point information of each potential sliding surface.

9. The adaptive individualized landslide prediction device according to claim 6, characterized in that: The calculation module comprises: A second model building unit is used to build a slope model according to the geometric feature data and the fault and joint information; A grid division unit, used for performing grid division through the slope model to obtain slope grid point data; A third model building unit is used to model the slope grid point data by discrete element method and input the strength parameter to obtain a sliding surface calculation model; A sample acquisition unit, used for inputting the polyline coordinate point information of each potential sliding surface into the sliding surface calculation model to obtain the calculation model sample corresponding to each potential sliding surface; A monitoring point selection unit is used to select monitoring points of the target slope; The second calculation unit is used to calculate the monitoring point displacement data of the slope surface from the initial state to the instability process of the monitoring point by the strength reduction method based on the calculation model sample, and the monitoring point displacement data includes the cumulative displacement, velocity, acceleration and displacement deflection angle of the monitoring point.

10. An adaptive individualized landslide prediction device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the adaptive individualized landslide prediction method according to any one of claims 1 to 5 when executing the computer program.

Images