Landslide monitoring data feature level fusion method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a landslide monitoring data feature level fusion method and device, electronic equipment and a storage medium. The method comprises the steps that landslide body deformation parameters are acquired; acquiring a plurality of influence factors related to the landslide; determining a key influence factor in the plurality of influence factors through clustering analysis; through stepwise regression analysis, taking the key influence factor as an independent variable and the landslide mass deformation parameter as a dependent variable, realizing feature level fusion, and obtaining a landslide mass surface displacement stepwise regression model; according to the method, the monitoring data can be effectively utilized, the monitored landslide deformation condition and the influence factors are comprehensively analyzed, the correlation of the influence factors is found out, an effective mathematical basis is provided for landslide deformation monitoring analysis, and the accuracy of landslide prediction and forecast is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data fusion, and in particular to a method, device, electronic equipment and storage medium for feature-level fusion of landslide monitoring data. Background Art

[0002] Landslide prediction is one of the key technologies for geological disaster prevention. It is based on multidisciplinary theories such as geology, statistics and information technology. Through continuous monitoring of the landslide body and its surrounding environment, key data such as displacement, rainfall, soil moisture, etc. can be collected. After processing and analysis, these data can reveal the laws and trends of landslide deformation. Landslide prediction technology uses mathematical models and algorithms to mine and interpret these data to predict the time, location and scale of possible landslides. This provides valuable early warning information for the power sector, enabling them to take timely and effective disaster reduction measures for power equipment installed on the mountain.

[0003] Landslide prediction analysis is a very important part of geological disaster monitoring and analysis. Different types of monitoring equipment on the landslide body simultaneously observe the deformation of the landslide and the surrounding conditions. How to effectively use these monitoring data is very important. In landslide monitoring, the displacement-time series collected by multiple sets of monitoring equipment are usually fused, which still fails to explore the correlation between landslide monitoring data and landslide prediction results. Summary of the invention

[0004] The main purpose of the present invention is to provide a method, device, electronic device and storage medium for feature-level fusion of landslide monitoring data, which can analyze the correlation of landslide influencing factors, provide an effective mathematical basis for landslide deformation monitoring and analysis, and further improve the accuracy of landslide prediction and forecasting.

[0005] To achieve the above-mentioned purpose, the present application provides a first aspect of a landslide monitoring data feature-level fusion method, the method comprising:

[0006] Obtaining deformation parameters of landslide body;

[0007] Obtain multiple influencing factors related to landslides;

[0008] Determining key influencing factors among the multiple influencing factors through cluster analysis;

[0009] Through stepwise regression analysis, the key influencing factors are used as independent variables, and the deformation parameters of the landslide body are used as dependent variables to achieve feature-level fusion and obtain a stepwise regression model of the landslide body surface displacement.

[0010] Optionally, obtaining deformation parameters of the landslide body includes:

[0011] For a landslide body provided with outdoor power grid equipment, the displacement size and moving speed of the landslide body are monitored at preset time intervals as deformation parameters of the landslide body.

[0012] Optionally, the obtaining of multiple influencing factors related to the landslide includes:

[0013] Acquire multiple candidate influencing factors related to the landslide, perform correlation analysis on each of the candidate influencing factors and the deformation parameter of the landslide body, and obtain a correlation coefficient between each of the candidate influencing factors and the landslide body;

[0014] The multiple influencing factors are preliminarily screened out from the candidate influencing factors according to the correlation coefficient.

[0015] Optionally, after preliminarily screening out the multiple influencing factors from the candidate influencing factors according to the correlation coefficient, the method further includes:

[0016] Obtaining specific data corresponding to the impact factor;

[0017] Determining the key influencing factors among the multiple influencing factors by cluster analysis includes:

[0018] After the specific data is standardized, the multiple influencing factors are clustered according to the specific data corresponding to each influencing factor to obtain a clustering result;

[0019] Determine an impact factor set according to the clustering result, each impact factor set corresponds to multiple impact factors, and the impact factors in different impact factor sets are different;

[0020] According to the set of influencing factors, using the trained landslide prediction model to obtain a landslide prediction result;

[0021] Comparing the actual landslide prediction result with the landslide prediction result output by the landslide prediction model to obtain the prediction accuracy corresponding to the influencing factor set;

[0022] By comparing the prediction accuracy rates corresponding to the influencing factor sets, the influencing factors included in the influencing factor set with the highest accuracy rate are determined as the key influencing factors.

[0023] Optionally, obtaining a landslide prediction result using a trained landslide prediction model according to the influencing factor set includes:

[0024] In the case where there are multiple influencing factor sets, the specific data of each influencing factor in each influencing factor set is input into the trained landslide prediction model to obtain the landslide prediction result;

[0025] By obtaining the specific data of each influencing factor in the same influencing factor set at multiple time points, multiple predictions are performed to obtain multiple landslide prediction results.

[0026] Optionally, the influencing factors include but are not limited to one or more of the following:

[0027] Geological conditions, topography, meteorological conditions, hydrological conditions, and human activities.

[0028] Optionally, the landslide surface displacement stepwise regression model includes hidden relationships among various monitoring data of the landslide, and the hidden relationships among various monitoring data of the landslide are expressed as the sum of the landslide surface displacement and the products of the various monitoring data and their corresponding weights.

[0029] A second aspect of the present application provides a device for fusion of landslide monitoring data at a feature level, comprising:

[0030] A data acquisition module is used to obtain deformation parameters of the landslide body;

[0031] Correlation analysis module, used to obtain multiple influencing factors related to landslide;

[0032] A cluster analysis module, used to determine key influencing factors among the multiple influencing factors through cluster analysis;

[0033] The stepwise regression analysis module is used to perform stepwise regression analysis, taking the key influencing factors as independent variables and the landslide deformation parameters as dependent variables, to achieve feature-level fusion and obtain a stepwise regression model of the landslide surface displacement.

[0034] A third aspect of the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the first aspect and any possible implementation thereof.

[0035] A fourth aspect of the present application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes each step of the method described in the first aspect.

[0036] The present application provides a method, device, electronic device and storage medium for characteristic-level fusion of landslide monitoring data, which obtains deformation parameters of a landslide body; obtains multiple influencing factors related to the landslide; determines key influencing factors among the multiple influencing factors through cluster analysis; uses stepwise regression analysis to take the key influencing factors as independent variables and the deformation parameters of the landslide body as dependent variables, realizes characteristic-level fusion, and obtains a stepwise regression model of the surface displacement of the landslide body; effectively utilizes monitoring data, comprehensively analyzes the deformation of the landslide obtained through monitoring and the influencing factors, finds out the correlation of the influencing factors, provides an effective mathematical basis for landslide deformation monitoring and analysis, and further improves the accuracy of landslide prediction and forecasting. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] in:

[0039] Figure 1 A schematic diagram of a flow chart of a method for feature-level fusion of landslide monitoring data provided in an embodiment of the present application;

[0040] Figure 2 A schematic diagram of the structure of a device for fusion of landslide monitoring data at a feature level provided in an embodiment of the present application;

[0041] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0043] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0044] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0045] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0046] See also Figure 1 , is a flow chart of a method for feature-level fusion of landslide monitoring data provided in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0047] 101. Obtain the deformation parameters of the landslide body.

[0048] The method in the embodiment of the present application can be executed on a landslide monitoring data feature level fusion device, and in a specific application can be implemented on an electronic device, and the above electronic device includes a terminal device such as a computer.

[0049] The landslide deformation parameters mentioned in the embodiments of the present application are characteristic indicators that describe the deformation of the landslide body under the action of internal and external forces. These parameters are crucial for understanding and predicting the stability and development trend of the landslide.

[0050] Optionally, the deformation parameters of the landslide body may mainly include the following aspects:

[0051] Displacement: The displacement of the landslide body in the horizontal and vertical directions is the most direct parameter to measure the activity of the landslide;

[0052] Speed: The rate of change of the displacement of the landslide body, which is an important indicator of the dynamic characteristics of the landslide and is used to assess the danger and urgency of the landslide;

[0053] Acceleration: describes the speed of landslide changes, which is of great significance for landslide risk identification and early warning;

[0054] Creep characteristics: The slow and continuous deformation of the landslide body under long-term action. This deformation is time-dependent and directly related to the stability of the landslide;

[0055] Deformation rate: The displacement change of the landslide body per unit time, which is used to monitor the activity of the landslide and predict the possibility of landslide occurrence;

[0056] Strain: The degree of deformation of the material inside the landslide body, which reflects the stress state and structural changes inside the landslide body;

[0057] Stress state: The stress distribution and magnitude inside the landslide body are closely related to the stability and deformation behavior of the landslide;

[0058] Crack development: the expansion and change of cracks on the surface and inside of the landslide body. The development of cracks can serve as a precursor to landslide activity.

[0059] Acoustic emission: High-frequency stress waves generated by the interaction between rock and soil particles during landslide deformation. The intensity and characteristics of AE signals can reflect the deformation characteristics of landslides.

[0060] By monitoring and analyzing these deformation parameters, the stability of the landslide body can be evaluated, the occurrence of landslides can be predicted, and a scientific basis can be provided for landslide prevention and control. In practical applications, these parameters are usually obtained through professional monitoring equipment and methods, such as GNSS monitoring, displacement meters, inclinometers, acoustic emission monitoring, etc.

[0061] Specifically, in the embodiments of the present application, for a landslide body equipped with outdoor power grid equipment, GPS, InSAR, radar interferometry and other technologies can be used to monitor the displacement and movement speed of the landslide body at preset time intervals as the above-mentioned landslide body deformation parameters.

[0062] 102. Obtain multiple influencing factors related to landslide.

[0063] The occurrence and development of landslides are affected by a variety of factors, which are summarized as influencing factors in the embodiments of the present application, mainly environmental factors.

[0064] Optionally, the above-mentioned influencing factors include but are not limited to one or more of the following:

[0065] Geological conditions, topography, meteorological conditions, hydrological conditions, and human activities.

[0066] In an optional implementation, the above step 102 includes:

[0067] Acquire multiple candidate influencing factors related to the landslide, perform correlation analysis on each of the candidate influencing factors and the deformation parameter of the landslide body, and obtain a correlation coefficient between each of the candidate influencing factors and the landslide body;

[0068] According to the above correlation coefficients, the above multiple influencing factors are preliminarily screened out from the above candidate influencing factors.

[0069] Specifically, multiple influencing factors related to landslides can be obtained, and then the correlation analysis between each influencing factor and the deformation parameters of the landslide body can be performed to obtain the correlation coefficient between each influencing factor and the landslide body. The larger the correlation coefficient value, the closer the degree of connection. According to the correlation coefficient, relevant influencing factors can be preliminarily screened out from multiple candidate factors, and specific data of the influencing shadow can be obtained, such as geological conditions (such as lithology, faults), topography (such as slope, slope direction), meteorological conditions (such as rainfall, temperature), hydrological conditions (such as groundwater level), human activities (such as engineering excavation, vegetation destruction) and other multi-source data.

[0070] 103. Determine the key influencing factors among the above multiple influencing factors through cluster analysis.

[0071] Cluster analysis is an unsupervised learning method that groups objects in a data set so that objects in the same group are more similar than objects in different groups. This technique is widely used in market analysis, customer segmentation, social network analysis, and other fields. The core idea of ​​cluster analysis is to group a collection of physical or abstract objects into multiple classes consisting of similar objects, usually based on the similarity of data.

[0072] In an optional implementation, after preliminarily screening out the plurality of influencing factors from the candidate influencing factors according to the correlation coefficient, the method further includes:

[0073] Obtain the specific data corresponding to the above impact factors;

[0074] The above step 103 includes:

[0075] After the above specific data are standardized, the above multiple influencing factors are clustered according to the specific data corresponding to each influencing factor to obtain a clustering result;

[0076] Determine an impact factor set according to the above clustering results, each impact factor set corresponds to multiple impact factors, and the impact factors in different impact factor sets are different;

[0077] According to the above-mentioned influencing factor set, the landslide prediction results are obtained using the trained landslide prediction model;

[0078] Compare the actual landslide prediction result with the landslide prediction result output by the landslide prediction model to obtain the prediction accuracy corresponding to the above-mentioned influencing factor set;

[0079] By comparing the prediction accuracy rates corresponding to the above influencing factor sets, the influencing factors included in the influencing factor set with the highest accuracy rate are determined as the above key influencing factors.

[0080] Further optionally, the landslide prediction result is obtained by using a trained landslide prediction model according to the above-mentioned influencing factor set, including:

[0081] In the case where there are multiple influencing factor sets, the specific data of each influencing factor in each influencing factor set is input into the trained landslide prediction model to obtain the landslide prediction result;

[0082] By obtaining the specific data of each influencing factor in the same influencing factor set at multiple time points, multiple predictions are performed to obtain multiple landslide prediction results.

[0083] For each landslide impact factor, the specific data of each landslide impact factor is determined, which can be called landslide impact factor data. Since different landslide impact factor data may have different dimensions and distributions, the data can be standardized first, including converting the data into a form with a mean of 0 and a standard deviation of 1; or scaling the data to a specific range, which is not limited in the embodiments of the present application.

[0084] Specifically, the multiple influencing factors can be clustered according to their respective multiple groups of landslide influencing factor data, and influencing factor sets of the multiple influencing factors can be determined according to the clustering results. Each influencing factor set corresponds to multiple influencing factors, and the influencing factors in different influencing factor sets are different.

[0085] In an optional implementation, during clustering, the data of each landslide influencing factor are first grouped according to a specified dimension. Specifically, the grouping can be based on spatial distribution. In practice, the data of each influencing factor in different spatial areas can be collected, such as landslide area 1, landslide area 2, ..., landslide area n. The landslide influencing factor data are then grouped according to the spatial distribution to obtain the data of each of the multiple influencing factors in each spatial distribution. Taking landslide area 2 as an example, the grouping result includes the landslide influencing factor data of each of the multiple influencing factors collected in landslide area 2.

[0086] Then, the landslide influencing factor data of the multiple influencing factors in each group are clustered in groups to obtain the clustering results of each group. By comparing the clustering results of multiple groups, multiple target landslide influencing factor sets are determined. Specifically, for the same type of set, such as the influencing factor set consisting of landslide influencing factor a, landslide influencing factor b and landslide influencing factor c, when the number of occurrences of the set in different groups reaches a threshold, it is determined as the target influencing factor set.

[0087] Further optionally, in the case of multiple target influencing factor sets, the landslide influencing factor data of each influencing factor in each target influencing factor set can be input into a trained landslide prediction model (such as a support vector machine, a random forest, a neural network, etc.) to obtain a landslide prediction result. By obtaining the data of each influencing factor in the same set at multiple time points, multiple predictions are performed to obtain multiple landslide prediction results. Then, the actual landslide prediction result is compared with the landslide prediction result output by the model to obtain the prediction accuracy corresponding to the target influencing factor set.

[0088] By comparing the prediction accuracies of multiple sets, the influencing factors contained in the set with the highest accuracy are determined as key influencing factors.

[0089] 104. Through stepwise regression analysis, the above key influencing factors are taken as independent variables, and the above landslide deformation parameters are taken as dependent variables to achieve feature-level fusion and obtain a stepwise regression model of landslide surface displacement.

[0090] In the embodiment of the present application, the purpose of stepwise regression analysis is to perform regression fitting analysis on the factors affecting landslide deformation screened by correlation analysis and cluster analysis, obtain the corresponding regression coefficients, and then calculate the final features and fusion results. In the stepwise regression analysis, the monitoring values ​​of each key influencing factor are used as independent variables, and the detected landslide deformation parameters are used as dependent variables to perform stepwise regression analysis.

[0091] In stepwise regression analysis, the optimal result of the model can be obtained by comparing the correlation coefficient, residual variance, F value, significance, etc. of different models.

[0092] Specifically, the above step 104 may include:

[0093] (1) Determine the preliminary regression model, taking all key landslide influencing factors as independent variables and the landslide deformation parameters as dependent variables;

[0094] (2) Gradually introduce variables: Start with the simplest model (i.e., only one independent variable) and gradually introduce other independent variables. In each step, the F value, significance level, and other indicators of the newly added variable are calculated to determine whether the variable significantly affects the dependent variable. If the newly added variable is not significant, it is removed from the model; if the added variable is significant, it is retained in the model and the next variable is introduced. Repeat the above steps until all significant variables are introduced into the model and no new significant variables can be added.

[0095] The specific mathematical implementation can refer to the following steps:

[0096] Step 1: For p regression independent variables X1, X2, …, X p Establish a univariate regression model with the dependent variable Y:

[0097] Y=β0+β i X i +e,i=1,…,p

[0098] Where e is the error term;

[0099] Calculate variable X i , the value of the F test statistic of the corresponding regression coefficient is recorded as Take the maximum value among them Right now

[0100]

[0101] For a given significance level α, the corresponding critical value is F (1) , Then X i1 Introduce the regression model and denote I1 as the set of selected variable indicators.

[0102] Step 2: Create dependent variable Y and independent variable subset {X i1 ,X1},...,{X i1 ,X i1-1}, {X i1 ,X i1+1}, ..., {X i1 ,X p} binary regression model (i.e. the regression variables of this regression model are binary), there are p-1 in total. Calculate the statistic value of the regression coefficient F test of the variable, denoted as Select the largest one and record it as The corresponding independent variable foot is marked as i2, that is:

[0103]

[0104] For a given significance level α, the corresponding critical value is F (2) , Then the variable X i2 Introduce the regression model. Otherwise, terminate the variable introduction process.

[0105] Step 3: Consider the dependent variable on the subset of variables {X i1 ,X i2 ,X k Repeat step 2 for the regression of}.

[0106] This method is repeated, each time selecting one independent variable from those that have not been introduced into the regression model, until no variables are introduced after verification.

[0107] Therefore, through the above-mentioned correlation analysis, cluster analysis and stepwise regression analysis, a stepwise regression model of the surface displacement of the landslide body can be obtained.

[0108] In an optional implementation, the landslide surface displacement stepwise regression model includes hidden relationships among various landslide monitoring data, and the hidden relationships among various landslide monitoring data are expressed as the sum of the landslide surface displacement and the product of the various monitoring data and their corresponding weights.

[0109] In some examples, the hidden relationship between various monitoring data of landslides can be written as:

[0110] The surface displacement of the landslide body = (GNSS monitoring point HF07 data × 0.587) + (displacement meter DCF11 data × 0.036) + (displacement meter DCF14 data × 0.519) - (displacement meter DCF15 data × 0.159) + (temperature × 0.028) + (humidity × 0.026) - (accumulated rainfall in the previous 48 hours × 0.010), and then the feature-level fusion results of clustering-stepwise regression analysis are obtained.

[0111] The above hidden relationship expression is the result of stepwise regression analysis, which shows the degree and direction of the influence of different monitoring data on the surface displacement of the landslide body. In this expression, the relationship between each monitoring data (such as GNSS monitoring point HF07 data, displacement meter DCF11 data, etc.) and the surface displacement of the landslide body is quantified by coefficients. These coefficients represent the relative importance and influence direction (positive or negative) of each monitoring data on the surface displacement of the landslide body. This hidden relationship expression is used to convert monitoring data into a quantitative assessment of landslide risk and provide important information for landslide management.

[0112] In the embodiment of the present application, feature-level fusion is achieved by combining clustering with stepwise regression analysis, and the monitored landslide deformation and influencing factors are comprehensively analyzed to explore the interaction between landslide deformation and various influencing factors, find out the correlation between environmental influencing factors, provide an effective mathematical basis for landslide deformation monitoring and analysis, and further improve the accuracy of landslide prediction and forecasting.

[0113] Based on the description of the above method embodiment, in one embodiment of the present application, a device for fusion of landslide monitoring data at feature level is also proposed. Figure 2 , Figure 2 The schematic diagram of the structure of a device for fusion of landslide monitoring data feature level provided in the embodiment of the present application is shown in FIG. Figure 2 As shown, the landslide monitoring data feature level fusion device 200 includes:

[0114] A data acquisition module 210 is used to acquire deformation parameters of the landslide body;

[0115] A correlation analysis module 220, used to obtain multiple influencing factors related to landslide;

[0116] A cluster analysis module 230, configured to determine key influencing factors among the multiple influencing factors through cluster analysis;

[0117] The stepwise regression analysis module 240 is used to perform stepwise regression analysis, taking the key influencing factors as independent variables and the landslide deformation parameters as dependent variables, to achieve feature-level fusion and obtain a stepwise regression model of the landslide surface displacement.

[0118] in, Figure 1 The method steps in the illustrated embodiment can be executed in the above-mentioned landslide monitoring data feature level fusion device 200, and will not be described in detail here.

[0119] Based on the description of the above method embodiment, in one embodiment of the present application, an electronic device is also provided. Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 3 As shown, the electronic device 300 includes a processor 301 and a memory 302, wherein the memory 302 stores a computer program. When the computer program is executed by the processor 301, the following operations are performed: Figure 1 Any step in the method embodiment shown. The electronic device 300 may also include an input / output device, etc. In a specific implementation, the electronic device may be a terminal device, etc.

[0120] In one embodiment, a computer-readable storage medium is further provided. The computer-readable storage medium stores a computer program. When the computer program is executed by the processor 301, the processor 301 executes any step in the above method embodiment.

[0121] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0122] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0123] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A feature-level fusion method for landslide monitoring data, characterized in that: The method comprises: Obtaining deformation parameters of landslide body; Obtain multiple influencing factors related to landslides; Determining key influencing factors among the multiple influencing factors through cluster analysis; Through stepwise regression analysis, the key influencing factors are used as independent variables, and the deformation parameters of the landslide body are used as dependent variables to achieve feature-level fusion and obtain a stepwise regression model of the landslide body surface displacement.

2. The method for feature-level fusion of landslide monitoring data according to claim 1, characterized in that: The step of obtaining deformation parameters of the landslide body includes: For a landslide body provided with outdoor power grid equipment, the displacement size and moving speed of the landslide body are monitored at preset time intervals as deformation parameters of the landslide body.

3. The method for characteristic-level fusion of landslide monitoring data according to claim 2 is characterized in that: The obtaining of multiple influencing factors related to landslide includes: Acquire multiple candidate influencing factors related to the landslide, perform correlation analysis on each of the candidate influencing factors and the deformation parameter of the landslide body, and obtain a correlation coefficient between each of the candidate influencing factors and the landslide body; The multiple influencing factors are preliminarily screened out from the candidate influencing factors according to the correlation coefficient.

4. The method for characteristic-level fusion of landslide monitoring data according to claim 3 is characterized in that: After preliminarily selecting the plurality of influencing factors from the candidate influencing factors according to the correlation coefficient, the method further includes: Obtaining specific data corresponding to the impact factor; Determining the key influencing factors among the multiple influencing factors by cluster analysis includes: After the specific data is standardized, the multiple influencing factors are clustered according to the specific data corresponding to each influencing factor to obtain a clustering result; Determine an impact factor set according to the clustering result, each impact factor set corresponds to multiple impact factors, and the impact factors in different impact factor sets are different; According to the set of influencing factors, using the trained landslide prediction model to obtain a landslide prediction result; Comparing the actual landslide prediction result with the landslide prediction result output by the landslide prediction model to obtain the prediction accuracy corresponding to the influencing factor set; By comparing the prediction accuracy rates corresponding to the influencing factor sets, the influencing factors included in the influencing factor set with the highest accuracy rate are determined as the key influencing factors.

5. The method for feature-level fusion of landslide monitoring data according to claim 4 is characterized in that: The step of obtaining a landslide prediction result using a trained landslide prediction model according to the set of influencing factors includes: In the case where there are multiple influencing factor sets, the specific data of each influencing factor in each influencing factor set is input into the trained landslide prediction model to obtain the landslide prediction result; By obtaining the specific data of each influencing factor in the same influencing factor set at multiple time points, multiple predictions are performed to obtain multiple landslide prediction results.

6. The method for feature-level fusion of landslide monitoring data according to claim 5 is characterized in that: The influencing factors include but are not limited to one or more of the following: Geological conditions, topography, meteorological conditions, hydrological conditions, and human activities.

7. The method for feature-level fusion of landslide monitoring data according to claim 1, characterized in that: The landslide surface displacement stepwise regression model includes hidden relationships among various monitoring data of the landslide, and the hidden relationships among various monitoring data of the landslide are expressed as the sum of the landslide surface displacement and the product of the various monitoring data and their corresponding weights.

8. A device for fusion of landslide monitoring data at feature level, characterized in that: include: A data acquisition module is used to obtain deformation parameters of the landslide body; Correlation analysis module, used to obtain multiple influencing factors related to landslide; A cluster analysis module, used to determine key influencing factors among the multiple influencing factors through cluster analysis; The stepwise regression analysis module is used to perform stepwise regression analysis, taking the key influencing factors as independent variables and the landslide deformation parameters as dependent variables, to achieve feature-level fusion and obtain a stepwise regression model of the landslide surface displacement.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

Images