Risk prediction methods, equipment and storage media

Abstract

This application provides a risk prediction method, device, and storage medium. The method includes: acquiring the index values ​​of a slope to be monitored under multiple target influence factors for a preset geological hazard; using a weighted radar chart method, based on the preset weights of the multiple target influence factors and the index values ​​of the slope to be monitored under these factors, predicting the risk of a preset geological hazard occurring on the slope to be monitored. This application considers the preset weights of the target influence factors and the index values ​​of the slope to be monitored under these factors, and uses a weighted radar chart method to predict the risk of geological hazard occurrence, thus improving the prediction accuracy.

Description

Technical Field

[0001] This application relates to the field of slope geological hazard risk assessment technology, and more specifically, to a risk prediction method, equipment, and storage medium. Background Technology

[0002] Slope unit risk prediction, also known as susceptibility prediction or sensitivity prediction, has the same ultimate goal: to solve the core problem of "which areas are most prone to geological disasters".

[0003] Among them, the research on slope geological hazard risk assessment needs to establish an analytical model to clearly describe the hazard body itself and make reasonable explanations for the geological hazard phenomena that have occurred or are occurring. At present, the combined weighting method is adopted, that is, subjective and objective predictions are made on the influencing factors of slope unit collapse and landslide, so as to assign subjective and objective weights to the influencing factors, and the subjective and objective weights are combined and weighted by additive synthesis, multiplicative synthesis, etc. to obtain the combined weight, and then the collapse and landslide potential of the slope unit is predicted based on the combined weight.

[0004] However, the accuracy of risk prediction for slope units using the above combined weighting method is insufficient. Summary of the Invention

[0005] In view of this, embodiments of this application provide a risk prediction method, device, and storage medium to solve the problem of insufficient accuracy in risk prediction for ramp units.

[0006] In a first aspect, embodiments of this application provide a risk prediction method, including:

[0007] Obtain the index values ​​of the slope to be monitored under multiple target influencing factors for preset geological hazards;

[0008] The weighted radar chart method is used to predict the risk of the preset geological disaster occurring on the slope to be monitored, based on the preset weights of the multiple target influence factors and the index values ​​of the slope to be monitored under the multiple target influence factors.

[0009] In an optional implementation, before predicting the risk of the preset geological hazard occurring on the slope to be monitored using the weighted radar chart method, based on the preset weights of the multiple target influence factors and the index values ​​of the slope to be monitored under the multiple target influence factors, the method further includes:

[0010] Obtain the subjective and objective influence weights of the multiple target influencing factors;

[0011] Based on the subjective influence weight and the objective influence weight, the preset weights of the multiple target influence factors are obtained.

[0012] In an optional implementation, before obtaining the subjective influence weights and objective influence weights of the plurality of target influence factors, the method further includes:

[0013] Identify multiple potential influencing factors for the occurrence of the preset geological disaster in the area where the slope to be monitored is located;

[0014] The index values ​​of multiple target slopes in the area where the pre-set geological hazard has occurred are obtained under multiple potential influencing factors, and the geological hazard kernel density of the multiple target slopes are obtained. The geological hazard kernel density of each target slope is the geological hazard kernel density of a local area including each target slope in the pre-divided area.

[0015] Based on the index values ​​of the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes, the multiple target influencing factors are determined from the multiple potential influencing factors.

[0016] In an optional implementation, determining the multiple target impact factors from the multiple potential impact factors based on the index values ​​under the multiple potential impact factors and the geological hazard kernel density of the multiple target slopes includes:

[0017] Based on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes, a first matrix data corresponding to the multiple target slopes is constructed;

[0018] The first matrix data is dimensionality reduced and denoised to obtain the second matrix data, which includes: index values ​​of the multiple potential influencing factors under multiple preset dimensions, and the geological hazard kernel density under the multiple preset dimensions;

[0019] Based on the index values ​​of the multiple potential influencing factors under the multiple preset dimensions and the geological hazard kernel density under the multiple preset dimensions, the multiple target influencing factors are determined from the multiple potential influencing factors.

[0020] In an optional implementation, determining the multiple target impact factors from the multiple potential impact factors based on the index values ​​of the multiple potential impact factors under the multiple preset dimensions and the geological hazard kernel density under the multiple preset dimensions includes:

[0021] Based on the index values ​​of the multiple potential influencing factors under the multiple preset dimensions, construct the factor vector of each potential influencing factor;

[0022] Based on the geological hazard kernel density under the multiple preset dimensions, a kernel density space vector is constructed;

[0023] Calculate the similarity between the factor vector and the kernel density space vector;

[0024] Based on the similarity, the target influence factors are determined from the plurality of potential influence factors.

[0025] In an optional implementation, determining the plurality of target influence factors from the plurality of potential influence factors based on the similarity includes:

[0026] Based on the similarity, the multiple potential influencing factors are classified to obtain the classification results of the multiple potential influencing factors, and the classification results are used to indicate the importance of each potential influencing factor;

[0027] Based on the classification results, the target impact factors are determined from the plurality of potential impact factors.

[0028] In an optional implementation, before constructing the first matrix data corresponding to the multiple target slopes based on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes, the method further includes:

[0029] The positive and negative influencing factors of the geological hazard are determined from the plurality of potential influencing factors;

[0030] The first range method and the second range method are used to process the index values ​​under the positive influence factor and the index values ​​under the negative influence factor, respectively, to obtain the standardized index values ​​of the positive influence factor and the standardized index values ​​of the negative influence factor.

[0031] The first matrix data corresponding to the multiple target slopes is constructed based on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes, including:

[0032] The first matrix data is constructed based on the standardized index values ​​of the positive influence factors, the standardized index values ​​of the negative influence factors, and the geological hazard kernel density of the multiple target slopes.

[0033] In an optional implementation, the step of dimensionality reduction and noise reduction of the first matrix data to obtain the second matrix data includes:

[0034] A correlation analysis was conducted on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes to obtain the analysis results;

[0035] If the analysis results meet the preset conditions, then principal component analysis is used to reduce the dimensionality and denoise the first matrix data to obtain the second matrix data.

[0036] Secondly, embodiments of this application also provide a risk prediction device, including:

[0037] The acquisition module is used to acquire the index values ​​of the slope to be monitored under multiple target influencing factors for preset geological hazards;

[0038] The prediction module is used to predict the risk of the preset geological disaster occurring on the slope to be monitored by using a weighted radar chart method, based on the preset weights of the multiple target influence factors and the index values ​​of the slope to be monitored under the multiple target influence factors.

[0039] In an optional implementation, the acquisition module is further configured to:

[0040] Obtain the subjective and objective influence weights of the multiple target influencing factors;

[0041] Based on the subjective influence weight and the objective influence weight, the preset weights of the multiple target influence factors are obtained.

[0042] In an optional implementation, the apparatus further includes:

[0043] The determination module is used to determine multiple potential influencing factors for the occurrence of the preset geological disaster in the area where the slope to be monitored is located;

[0044] The acquisition module is also used to acquire the index values ​​of multiple target slopes in the area where the preset geological disaster has occurred under the multiple potential influencing factors, and the geological disaster kernel density of the multiple target slopes. The geological disaster kernel density of each target slope is the geological disaster kernel density of a local area including each target slope in the pre-divided area.

[0045] The determining module is further configured to determine the multiple target impact factors from the multiple potential impact factors based on the index values ​​under the multiple potential impact factors and the geological hazard kernel density of the multiple target slopes.

[0046] In an optional implementation, the determining module is specifically used for:

[0047] Based on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes, a first matrix data corresponding to the multiple target slopes is constructed;

[0048] The first matrix data is dimensionality reduced and denoised to obtain the second matrix data, which includes: index values ​​of the multiple potential influencing factors under multiple preset dimensions, and the geological hazard kernel density under the multiple preset dimensions;

[0049] Based on the index values ​​of the multiple potential influencing factors under the multiple preset dimensions and the geological hazard kernel density under the multiple preset dimensions, the multiple target influencing factors are determined from the multiple potential influencing factors.

[0050] In an optional implementation, the determining module is specifically used for:

[0051] Based on the index values ​​of the multiple potential influencing factors under the multiple preset dimensions, construct the factor vector of each potential influencing factor;

[0052] Based on the geological hazard kernel density under the multiple preset dimensions, a kernel density space vector is constructed;

[0053] Calculate the similarity between the factor vector and the kernel density space vector;

[0054] Based on the similarity, the target influence factors are determined from the plurality of potential influence factors.

[0055] In an optional implementation, the determining module is specifically used for:

[0056] Based on the similarity, the multiple potential influencing factors are classified to obtain the classification results of the multiple potential influencing factors, and the classification results are used to indicate the importance of each potential influencing factor;

[0057] Based on the classification results, the target impact factors are determined from the plurality of potential impact factors.

[0058] In an optional implementation, the determining module is further configured to:

[0059] The positive and negative influencing factors of the geological hazard are determined from the plurality of potential influencing factors;

[0060] The device further includes:

[0061] The processing module is used to process the index values ​​under the positive influence factor and the index values ​​under the negative influence factor using the first range method and the second range method, respectively, to obtain the standardized index values ​​of the positive influence factor and the standardized index values ​​of the negative influence factor.

[0062] The determining module is specifically used for:

[0063] The first matrix data is constructed based on the standardized index values ​​of the positive influence factors, the standardized index values ​​of the negative influence factors, and the geological hazard kernel density of the multiple target slopes.

[0064] In an optional implementation, the determining module is further configured to:

[0065] A correlation analysis was conducted on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes to obtain the analysis results;

[0066] If the analysis results meet the preset conditions, then principal component analysis is used to reduce the dimensionality and denoise the first matrix data to obtain the second matrix data.

[0067] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform any of the risk prediction methods described in the first aspect.

[0068] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the risk prediction method described in any of the first aspects.

[0069] This application provides a risk prediction method, device, and storage medium. The method includes: acquiring the index values ​​of a slope to be monitored under multiple target influence factors for a preset geological hazard; using a weighted radar chart method, based on the preset weights of the multiple target influence factors and the index values ​​of the slope to be monitored under these factors, predicting the risk of a preset geological hazard occurring on the slope to be monitored. This application considers the preset weights of the target influence factors and the index values ​​of the slope to be monitored under these factors, and uses a weighted radar chart method to predict the risk of geological hazard occurrence, thus improving the prediction accuracy.

[0070] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0071] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0072] Figure 1 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 1 ;

[0073] Figure 2A schematic diagram of the weighted radar chart model provided in the embodiments of this application;

[0074] Figure 3 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 2 ;

[0075] Figure 4 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 3 ;

[0076] Figure 5 A schematic diagram illustrating potential influencing factors provided for embodiments of this application;

[0077] Figure 6 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 4 ;

[0078] Figure 7 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 5 ;

[0079] Figure 8 This is a schematic diagram of the structure of the risk prediction device provided in the embodiments of this application;

[0080] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0081] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0082] Because geological hazards are highly complex and pose a threat to society, it is difficult to accurately identify the key influencing factors (i.e., evaluation factors) that describe their inherent changes. Furthermore, the hazard itself generates other factors that can express its properties, and these factors can manifest in different forms depending on the region. For example, slope, slope variability, and profile curvature are relevant. Slope variability simply refers to the change in ground slope, while profile curvature refers to the change in ground slope along the direction of maximum gradient. Therefore, slope variability and profile curvature are extensions of slope at a deeper level. In conclusion, research on slope geological hazard risk assessment first requires establishing an analytical model to clearly describe the hazard itself, helping to clarify the logical relationships of hazard development or disaster, and providing reasonable explanations for past or ongoing geological hazard phenomena. This necessitates establishing a comprehensive, reasonable, and standardized evaluation index system.

[0083] This shows that selecting appropriate key influencing factors, weighting methods, and suitable evaluation methods is crucial for assessing slope landslide potential, which in turn affects the quality and even the success or failure of subsequent slope geological hazard risk management.

[0084] Regarding the existing combined weighting methods, additive and multiplicative methods are subject to a lot of subjective judgment factors and it is difficult to guarantee the accuracy of weight allocation. The superior-inferior solution distance method (TOPSIS method) has a better effect in balancing the weights of "one subject and one object". However, there is currently a lack of effective combined weighting methods that can be applied to a variety of different subjective and objective objectives. In other words, the accuracy of risk prediction using both combined weighting methods and superior-inferior solution distance method is insufficient.

[0085] Based on this, this application provides a risk prediction method based on the combined weighting method and the weighted radar chart method. It can select target influencing factors from a large number of potential influencing factors affecting the preset geological disasters, and then establish an objective combined weight by comprehensively weighting the target influencing factors. Finally, the evaluation model of the weighted radar chart method is used to quantitatively evaluate the risk of the preset geological disasters on the slope, so as to predict the risk of the preset geological disasters on the slope, thereby improving the prediction accuracy.

[0086] The risk prediction method provided in this application will be described below with reference to several specific embodiments.

[0087] Figure 1 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 1 In this embodiment, the executing entity can be an electronic device, such as a terminal or a server.

[0088] like Figure 1 As shown, the method may include:

[0089] S101. Obtain the index values ​​of the slope to be monitored under multiple target influencing factors for preset geological hazards.

[0090] The preset geological hazards can be geological hazards such as collapses and landslides. The target influencing factor for the preset geological hazards can be the effective influencing factor that causes the preset geological hazards to occur on the slope. That is, the target influencing factor can be selected from a large number of potential influencing factors for slope collapse, landslide and instability.

[0091] Among them, the slope to be monitored has one index value under one target influencing factor. Multiple target influencing factors may include, for example, topographic humidity index, surface roughness, topographic relief, etc.

[0092] S102. Using the weighted radar chart method, based on the preset weights of multiple target influence factors and the index values ​​of the slope to be monitored under multiple target influence factors, the risk of the slope to be monitored experiencing preset geological disasters is predicted.

[0093] The preset weights of the target impact factors are used to indicate the degree of influence of the target impact factors on the preset geological hazards. For example, if the degree of influence of the topographic humidity index on the preset geological hazards is higher than that of the topographic relief degree on the preset geological hazards, then the preset weight of the topographic humidity index is higher than that of the topographic relief degree.

[0094] Among them, the weighted radar chart method corresponds to the comprehensive evaluation function of the weighted radar chart model. By adopting the comprehensive evaluation function of the weighted radar chart method, risk prediction is carried out based on the preset weights of multiple target influence factors and the index values ​​of the slope to be monitored under multiple target influence factors. The risk value of the slope to be monitored for the occurrence of the preset geological disaster is obtained, indicating the possibility of the slope to be monitored for the occurrence of the preset geological disaster.

[0095] For example, if the preset geological hazard is landslide, then the preset risk value of the geological hazard can be the landslide instability potential value.

[0096] In some embodiments, the risk level of a pre-set geological hazard on the slope to be monitored can be determined based on the risk value of the pre-set geological hazard on the slope to be monitored. The risk level can include four levels: extremely high potential, high potential, medium potential, and low potential.

[0097] Referring to Table 1, which is a classification table of slope landslide potential, the slope landslide potential levels include: extremely high potential, high potential, medium potential, and low potential, with corresponding landslide instability potential values ​​of λ≥0.6151, 0.6151>λ≥0.5205, 0.5205>λ≥0.3990, and 0.3990>λ, respectively. In addition, different colors can be used to mark the slopes to be monitored on the electronic map of the area where the slopes to be monitored are located, according to the slope landslide potential level.

[0098]

[0099] Table 1

[0100] It is worth noting that ArcGIS software was used to assign the calculated instability potential values ​​to the respective slopes. This software also provides grading methods such as equal intervals, quantiles, natural breaks, geometric intervals, and standard deviations. These grading methods can be used to classify the landslide potential values ​​of the slopes. Since there may be target slopes in the area where the monitored slopes are located that have already experienced predetermined geological hazards, the grading results of the target slopes and their actual risk levels can be compared to determine the final grading method. Experiments showed that the grading results of the natural break method generally conformed to the hazard-prone characteristics of the slopes in the region and had a high degree of agreement with previous research results in the region, showing the best effect. In addition, the natural break itself is a high-quality boundary for grading and has strong statistical significance, with obvious advantages. Therefore, the natural break method can be selected as the grading method for risk levels.

[0101] In the risk prediction method of this embodiment, the index values ​​of the slope to be monitored under multiple target influence factors for preset geological hazards are obtained. A weighted radar chart method is used to predict the risk of the preset geological hazard occurring on the slope to be monitored, based on the preset weights of the multiple target influence factors and the index values ​​of the slope under these factors. By considering the preset weights of the target influence factors and the index values ​​of the slope under these factors, and employing the weighted radar chart method to predict the risk of geological hazard occurrence, the prediction accuracy is improved.

[0102] Figure 2 This is a schematic diagram of the weighted radar chart model provided in the embodiments of this application, as shown below. Figure 2 As shown, let F i The weight vector is obtained by normalizing the preset weights of multiple target influence factors in descending order, where i = 1, 2, ..., m, m represents the number of multiple target influence factors, and the corresponding central angle θ of the sector is... i =2πF i Specifically, it includes the following steps:

[0103] (1) Draw a unit circle and draw an eastward ray OH1 through the center O. The right side is east. The unit circle has markings of 0.2, 0.4, 0.6, 0.8 and 1.0.

[0104] (2) Rotate θ clockwise along ray OH1 sequentially i =2πF i From the angle, draw the remaining m-1 rays, namely rays OH2, OH3, ..., OH m .

[0105] (3) Construct ∠H1OH2, ∠H2OH3, ..., ∠H m-1 OH m , ∠H m OH1's diagonal is OK i .

[0106] (4) OK i As the indicator axis, the indicator χ=(x1,x2,...,x m Mark the corresponding point P on the axis according to its length. i Connect P1, P2, ..., P in sequence. m Let the polygons P1P2...P be represented by... m The area S = s1 + s2 + ... + s m The perimeter is C = l1 + l2 + ... + l m The weighted radar chart model is obtained, x m The preset weight is the weight of the m-th target influence factor.

[0107] The process of establishing the comprehensive evaluation function of the weighted radar chart model is as follows:

[0108] (1) Construct the evaluation vector λ using the area S and perimeter C of the weighted radar chart model as feature vectors. j =(λ js ,λ jc The calculation formula is as follows:

[0109]

[0110]

[0111] In the formula: λ js S represents the area of ​​the j-th evaluation object. j The ratio of the area of ​​the largest evaluated object to the area of ​​the largest evaluated object reflects the strength or weakness of that object; λ jc S represents the area of ​​the j-th evaluation object. j Area of ​​a circle with the same perimeter The ratio reflects the balanced development degree of the target influencing factors of the evaluation object. Here, the j-th evaluation object is the j-th slope in the area where the slope to be monitored is located. In other words, when using the weighted radar chart method to predict the risk of the slope to be monitored, it is necessary to follow... Figure 2 The method shown constructs the area and perimeter of polygons corresponding to each slope in the region to be monitored. The area of ​​the largest evaluation object is the area of ​​the largest polygon corresponding to each slope in the region to be monitored. The area of ​​a circle with the same perimeter is the area of ​​a circle with the same perimeter as the j-th slope. The perimeter of the j-th slope is C. j .

[0112] (2) Define the expression for the comprehensive evaluation function of the weighted radar chart model as follows:

[0113]

[0114] Where, λ j The value represents the potential for landslides. The magnitude of the potential for landslides can be used to compare the degree of landslide instability. The larger the value, the easier it is for landslide instability to occur, and the smaller the value, the less likely it is for landslide instability to occur.

[0115] Figure 3 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 2 ,like Figure 3 As shown, before predicting the risk of a predetermined geological hazard on the slope to be monitored using the weighted radar chart method, based on the preset weights of multiple target influence factors and the index values ​​of the slope under these factors, this method may further include:

[0116] S201. Obtain the subjective and objective influence weights of multiple target influence factors.

[0117] Among them, subjective influence weight can be the influence weight assigned to the target influence factor using subjective weighting method, and objective influence weight can be the influence weight assigned to the target influence factor using objective weighting method. Subjective weighting method can include, for example, fuzzy comprehensive evaluation, expert experience method, etc., while objective weighting method can include, for example, principal component analysis method, entropy method, and CRITIC method, etc.

[0118] S202. Based on the subjective influence weight and the objective influence weight, obtain the preset weights of multiple target influence factors.

[0119] After obtaining the subjective and objective impact weights of multiple target impact factors, the multiple target impact factors can be combined and weighted according to these subjective and objective impact weights to obtain the combined weight of the multiple target impact factors. This combined weight is a preset weight.

[0120] Multiple target influencing factors may include: topographic humidity index, land use type, normalized vegetation index, fault distance, surface incision depth, surface roughness, river distance (distance between slope and river), road distance (distance between slope and road), topographic relief, annual average rainfall, and plane curvature. Calculate the combined weight of each target influencing factor, and rank the multiple target influencing factors according to the combined weight. Refer to Table 2, which is a ranking table of combined weight values.

[0121] serial number Target Impact Factor Combined weight values Sort k1 Topographic humidity index 0.1376 1 k2 Land use types 0.1228 2 k3 Normalized Difference Vegetation Index 0.0956 3 k4 Fault distance 0.0934 4 k5 Surface cutting depth 0.0914 5 k6 Surface roughness 0.0862 6 k7 River distance 0.0859 7 k8 Road distance 0.0752 8 k9 Topographic relief 0.0731 9 k10 Annual average rainfall 0.0721 10 k11 Plane curvature 0.0668 11

[0122] Table 2

[0123] It is worth noting that the minimum deviation method can be used to calculate the combined weight of the target impact factors based on the subjective and objective impact weights. The principle of the minimum deviation method is as follows:

[0124] Assuming that m independent weighting methods are used (including x subjective weighting methods and y objective weighting methods, m = x + y) and n target influence factors are selected, then the weight vector matrix after combining the i-th weighting methods is D. i =(d i1 d i2 , ..., d in ) T ,in, d i1 Let j be the influence weight assigned to the first target influence factor using the i-th weighting method, and j be the j-th target influence factor, with a value ranging from 1 to n.

[0125] Define a single-objective optimization model based on a deviation function to minimize the total deviation Z of each weighting method, and calculate the comprehensive optimal combination coefficient vector K = (ε1, ε2, ..., ε) of each objective influence factor when the deviation is minimized under each weighting method. m ) T , superimposed D i We obtain D, and the combined weight vector of each objective influence factor is F = DK.

[0126] The calculation formula for the single-objective optimization model based on the deviation function is as follows:

[0127]

[0128] In the formula: d ij The standardized influence weight of the j-th target influence factor index after the i-th weighting method.

[0129] Using the principle of the minimum deviation method described above, we construct the Lagrangian function L(d,δ) for m = 1, 2, 3, 4, 5 (assuming m = 5) and find the extreme values, obtaining the following system of equations:

[0130]

[0131] In the formula: δ refers to the Lagrange multiplier.

[0132] The subjective and objective weights d obtained by different methods ij Substituting the specific values ​​into the above system of equations, we can see by Cramer's rule that this system of equations has a unique non-zero solution. After solving, we obtain the comprehensive optimal combination coefficient vector K = (ε1, ε2, ε3, ε4, ε5). T The final combined weight vector is F = DK, where each element in the combined weight vector corresponds to the combined weight of a target influence factor.

[0133] Figure 4 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 3 ,like Figure 4 As shown, before obtaining the subjective and objective influence weights of multiple target influence factors, the method may further include:

[0134] S301. Identify multiple potential influencing factors for the occurrence of preset geological disasters in the area where the slope to be monitored is located.

[0135] Among them, an impact factor system for the occurrence of pre-set geological disasters in the area where the slope to be monitored is established. After risk prediction of the slope to be monitored, a more comprehensive and scientific multi-potential impact factor analysis system can be constructed from the perspective of existing fact analysis, including: topography, geological conditions, human engineering and environment.

[0136] Figure 5 A schematic diagram of potential influencing factors provided for embodiments of this application, such as... Figure 5 As shown, the categories include topography, geological conditions, human engineering, and environment.

[0137] Potential influencing factors for topography and geomorphology may include: elevation, slope, aspect, slope variability, aspect variability, plan curvature, profile curvature, surface roughness, topographic relief, surface incision depth, topographic moisture index, and sediment transport index.

[0138] Potential influencing factors related to geological conditions may include: engineering geological rock groups, slope structure, and fault distance, etc.

[0139] Potential influencing factors for human engineering projects may include: distance from residence and distance from roads, etc.

[0140] Potential environmental influencing factors may include: vegetation index, land use type, average annual rainfall, and river distance, etc.

[0141] S302. Obtain the index values ​​of multiple target slopes in the area where the monitored slope has experienced preset geological disasters under multiple potential influencing factors, and the geological disaster kernel density of multiple target slopes.

[0142] Among them, there are multiple target slopes in the area where the slope to be monitored has experienced preset geological disasters. The index values ​​of these multiple target slopes under multiple potential influencing factors and the geological disaster kernel density of the multiple target slopes are obtained. The geological disaster kernel density of each target slope is the geological disaster kernel density of the local area including each target slope in the pre-divided area. The area where the slope to be monitored is pre-divided into multiple local areas, and each local area includes at least one slope. The geological disaster kernel density of each target slope is the ratio of the number of preset geological disasters that have occurred in the local area where the target slope is located to the area of ​​the local area where the target slope is located.

[0143] S303. Based on the index values ​​of multiple potential influencing factors and the geological hazard kernel density of multiple target slopes, determine multiple target influencing factors from multiple potential influencing factors.

[0144] Based on the index values ​​of multiple potential influencing factors and the geological hazard kernel density of multiple target slopes, multiple potential influencing factors can be screened to determine multiple target influencing factors. In other words, when screening target influencing factors, the index and kernel density of target slopes in the area where the slope to be monitored has already experienced preset geological hazards are taken into consideration.

[0145] It is worth noting that the kernel density analysis tool in ArcGIS can be used to obtain the geological hazard kernel density classification map of Baxu County in a certain region. Based on the raster value extraction function of GIS, 21 potential influencing factors and kernel densities can be extracted onto the disaster-causing area of ​​the geological hazard point (i.e. the target slope where the preset geological hazard has occurred), resulting in a 150*22 raw data (Table 1). Refer to Table 3, which is a table of index values ​​and kernel densities of the target slope under multiple potential influencing factors. The hazard name is the name of the target slope where the preset geological hazard has occurred.

[0146]

[0147] Table 3

[0148] In one possible implementation of step S303 above, multiple target impact factors are determined from the multiple potential impact factors based on the index values ​​under multiple potential impact factors and the geological hazard kernel density of multiple target slopes, including:

[0149] Based on the index values ​​of multiple potential influencing factors and the geological hazard kernel density of multiple target slopes, a first matrix data corresponding to multiple target slopes is constructed; the first matrix data is dimensionality reduced and denoised to obtain a second matrix data, as well as the geological hazard kernel density under multiple preset dimensions; based on the index values ​​of multiple potential influencing factors under multiple preset dimensions and the geological hazard kernel density under multiple preset dimensions, multiple target influencing factors are determined from multiple potential influencing factors.

[0150] Based on the index values ​​of each target slope under multiple potential influencing factors and the geological hazard kernel density of multiple target slopes, a first matrix data corresponding to multiple target slopes can be constructed. The elements in the first matrix data include the index values ​​of the target slope under multiple potential influencing factors and the geological hazard kernel density of multiple target slopes. Referring to Table 3, the first matrix data can be a 150*22 matrix data.

[0151] Then, the first matrix data is dimensionality reduced and denoised to obtain the second matrix data. The second matrix data includes the index values ​​of multiple potential influencing factors under multiple preset dimensions. The second matrix data is a low-dimensional principal component matrix table. Each potential influencing factor under a preset dimension has one index value. Taking 7 preset dimensions as an example, refer to Table 4. Table 4 shows the index values ​​and kernel density of potential influencing factors under the 7 preset dimensions corresponding to the second matrix data.

[0152] In an optional implementation, the first matrix data is dimensionality reduced and denoised to obtain the second matrix data, including: performing correlation analysis on the index values ​​under multiple potential influencing factors and the geological hazard kernel density of multiple target slopes to obtain the analysis results; if the analysis results meet the preset conditions, then the principal component analysis method is used to dimensionality reduce and denoise the first matrix data to obtain the second matrix data.

[0153] A correlation analysis was conducted on the index values ​​of multiple target slopes under multiple potential influencing factors and the kernel density of geological hazards on multiple target slopes. The analysis results were obtained. Under the condition that the analysis results met the preset conditions, principal component analysis was used to reduce the dimension and denoise of the first matrix data to obtain the second matrix data. The correlation analysis can be performed using SPSS software. The analysis results are used to indicate the correlation between the index values ​​of multiple potential influencing factors and the kernel density of geological hazards on multiple target slopes. The analysis results can include the Kaiser-Meyer-Olkin (KMO) test value and Bartlett's test value. The KMO test value can be 0.6.10 and the Bartlett's test value can be 0.000. The preset conditions can include, for example, KMO test value > 0.5 and Bartlett's test value < 0.05.

[0154] Then, based on the index values ​​of multiple potential influencing factors under multiple preset dimensions and the kernel density of geological hazards under multiple preset dimensions, multiple potential influencing factors are screened to determine multiple target influencing factors. In other words, when screening target influencing factors, the index values ​​of multiple potential influencing factors under multiple preset dimensions and the kernel density of geological hazards under multiple preset dimensions are taken into consideration. The index values ​​and kernel densities under the preset dimensions obtained by dimensionality reduction and noise reduction are used for screening, saving information space and improving processing efficiency.

[0155]

[0156] Table 4

[0157] As shown in Table 4, the feature values ​​of these 7 preset dimensions are all greater than 1, and the cumulative information contribution rate is 67.06%, which is close to 70%, and can reflect the environmental characteristics of disaster formation and disaster.

[0158] Figure 6 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 4 ,like Figure 6 As shown, based on the index values ​​of multiple potential influencing factors under multiple preset dimensions and the geological hazard kernel density under multiple preset dimensions, multiple target influencing factors are determined from the multiple potential influencing factors, including:

[0159] S401. Construct factor vectors for each potential influencing factor based on the index values ​​of multiple potential influencing factors under multiple preset dimensions.

[0160] Based on the index values ​​of multiple potential influencing factors under multiple preset dimensions, a factor vector is constructed for each potential influencing factor. Each potential influencing factor corresponds to a factor vector, and the elements in the factor vector of each potential influencing factor are the index values ​​of each original influencing shadow under multiple preset dimensions.

[0161] S402. Construct a kernel density spatial vector based on the kernel density of geological hazards under multiple preset dimensions.

[0162] A kernel density space vector is constructed based on the kernel density of geological hazards under multiple preset dimensions. The elements in the kernel density space vector are the kernel densities of geological hazards under multiple preset dimensions.

[0163] S403. Calculate the similarity between the factor vector and the kernel density space vector.

[0164] S404. Based on similarity, identify multiple target impact factors from multiple potential impact factors.

[0165] Calculate the similarity between the factor vector and the kernel density space vector of each potential influencing factor, and based on this similarity, select the potential influencing factors whose similarity exceeds a preset similarity threshold as target influencing factors. In other words, the similarity between the target influencing factor and the kernel density space vector exceeds the preset similarity threshold.

[0166] Referring to Table 5, which is a similarity ranking table between factor vectors and kernel density space vectors, the cosine value is the cosine similarity between factor vectors and kernel density space vectors, and the angle value is the angle between factor vectors and kernel density space vectors.

[0167]

[0168]

[0169] Table 5

[0170] In one possible implementation of step S404 above, determining multiple target impact factors from multiple potential impact factors based on similarity includes: classifying multiple potential impact factors according to similarity to obtain classification results of multiple potential impact factors, the classification results being used to indicate the importance of each potential impact factor; and determining multiple target impact factors from multiple potential impact factors based on the importance.

[0171] Based on the similarity between the factor vector and the kernel density space vector, multiple potential influencing factors are classified, resulting in classification results for each potential influencing factor. The classification results indicate the importance of each potential influencing factor. Different similarity ranges correspond to different classification results. The classification result corresponding to the similarity range to which the potential influencing factor falls is taken as the classification result of that potential influencing factor.

[0172] Then, based on the grading results, multiple target impact factors are determined from multiple potential impact factors. The grading results of the target impact factors can be preset grading results in the grading results. For example, the grading results can include major, minor and general, and the preset grading results can include major and minor.

[0173] Refer to Table 6, which shows the importance of potential influencing factors.

[0174]

[0175]

[0176] Table 6

[0177] It is worth noting that the improved cosine similarity is used to calculate the angle between the factor vector and the kernel density space vector. The improved cosine similarity calculation formula is as follows:

[0178]

[0179] In the formula: d ij ρ represents the value of the i-th potential influence factor in the j-th dimension. j Represents the j-th dimension

[0180] The closer the relationship between the formation of a geological disaster body and the disaster itself, the less close the relationship between the two.

[0181] Figure 7 A flowchart illustrating the risk prediction method provided in this application embodiment. Figure 5 ,like Figure 7 As shown, before constructing the first matrix data corresponding to multiple target slopes based on the index values ​​under multiple potential influencing factors and the geological hazard kernel density of multiple target slopes, the method may further include:

[0182] S501. Identify the positive and negative influencing factors of geological hazards from multiple potential influencing factors.

[0183] Among them, positive influencing factors are potential influencing factors that are positively correlated with the preset geological hazards. For example, the steeper the slope, the more likely the preset geological hazard will occur. Negative influencing factors are potential influencing factors that are negatively correlated with the preset geological hazards. For example, the smaller the surface roughness, the smaller the normalized vegetation index, the smaller the engineering geological rock group, the smaller the slope structure, the smaller the fault distance, the smaller the residential distance, the smaller the road distance, and the smaller the river distance, the less likely the preset geological hazard will occur.

[0184] S502. Using the first range method and the second range method, the index values ​​under the positive influence factor and the index values ​​under the negative influence factor are processed respectively to obtain the standardized index values ​​of the positive influence factor and the standardized index values ​​of the negative influence factor.

[0185] To eliminate the correlation differences between negative and positive influencing factors and preset geological hazards, and to reduce the amount of data, the range method can be used to standardize the index values ​​under positive and negative influencing factors.

[0186] The first range method is used to process the index values ​​of each target slope under the positive influence factor to obtain the standardized index values ​​of each target slope under the positive influence factor. The second range method is used to process the index values ​​of each target slope under the negative influence factor to obtain the standardized index values ​​of each target slope under the negative influence factor.

[0187] The range method was used to standardize the values ​​of each potential influencing factor and the kernel density index of slope geological hazards. The formula for the range method is as follows:

[0188]

[0189] Where: β i Let α be the standardized index value of the i-th target slope, including positive indices (i.e., standardized index values ​​of each target slope under positive influence factors) and negative indices (i.e., index values ​​of each target slope under negative influence factors). i Let max(α) be the index value of the i-th target slope under the potential influencing factors. i ) represents the maximum value of the index for all target slopes under the potential influencing factors, min(α) i ) represents the minimum index value among all target slopes under the potential influencing factors.

[0190] Using the above formula, the positive index (i.e., the standardized index value of each target slope under the positive influence factor) and the negative index (i.e., the index value of each target slope under the negative influence factor) are standardized to between 0 and 1.

[0191] Of course, it is understandable that in step S102 above, the range method can also be used to standardize the index values ​​of the slope to be monitored under multiple target influencing factors for preset geological hazards. The specific processing method is the same as... Figure 7 Similarly, I will not go into details here.

[0192] In an optional implementation, based on the index values ​​of multiple potential influencing factors and the geological hazard kernel density of multiple target slopes, a first matrix of data corresponding to multiple target slopes is constructed, including:

[0193] The first matrix data is constructed based on the standardized index values ​​of positive and negative impact factors and the geological hazard kernel density of multiple target slopes.

[0194] Based on the standardized index values ​​of positive and negative impact factors and the geological hazard kernel density of multiple target slopes, a first matrix data is constructed. The elements in the first matrix data include the standardized index values ​​of positive and negative impact factors for each target slope and the geological hazard kernel density of multiple target slopes.

[0195] In summary, the embodiments of this application utilize principal component analysis, improved cosine similarity, and minimum deviation method to measure the criticality of slope landslide instability behavior with various potential influencing factors and obtain their combined weights. This enables the establishment of an objective and reasonable criterion for slope landslide instability under the influence of multiple potential factors, overcoming the differences between various potential influencing factors. In addition, the use of a weighted radar chart model suitable for the combined weight method better reflects the relative strength of individual slope landslide instability and also takes into account the stability between various target influencing factors, providing a new method for slope geological hazard risk assessment.

[0196] Based on the same inventive concept, this application also provides a risk prediction device corresponding to the risk prediction method. Since the principle of the device in this application is similar to the risk prediction method described above in this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0197] Figure 8 This is a schematic diagram of the risk prediction device provided in an embodiment of this application. This device can be integrated into an electronic device. Figure 8 As shown, the device may include:

[0198] The acquisition module 601 is used to acquire the index values ​​of the slope to be monitored under multiple target influencing factors for preset geological hazards;

[0199] The prediction module 602 is used to predict the risk of a preset geological disaster on the slope to be monitored by using the weighted radar chart method, based on the preset weights of multiple target influence factors and the index values ​​of the slope to be monitored under multiple target influence factors.

[0200] In an optional implementation, the acquisition module 601 is further configured to:

[0201] Obtain the subjective and objective influence weights of multiple target influencing factors;

[0202] Based on the subjective influence weight and the objective influence weight, the preset weights of multiple target influence factors are obtained.

[0203] In an optional embodiment, the device further includes:

[0204] The determination module 603 is used to determine multiple potential influencing factors for the occurrence of preset geological disasters in the area where the slope to be monitored is located;

[0205] The acquisition module 601 is also used to acquire the index values ​​of multiple target slopes in the area where the slope to be monitored has experienced preset geological disasters under multiple potential influencing factors, and the geological disaster kernel density of multiple target slopes. The geological disaster kernel density of each target slope is the geological disaster kernel density of the local area including each target slope in the pre-divided area.

[0206] The determination module 603 is also used to determine multiple target impact factors from multiple potential impact factors based on the index values ​​under multiple potential impact factors and the geological hazard kernel density of multiple target slopes.

[0207] In an optional implementation, the determining module 603 is specifically used for:

[0208] Based on the index values ​​of multiple potential influencing factors and the geological hazard kernel density of multiple target slopes, a first matrix of data corresponding to multiple target slopes is constructed.

[0209] The first matrix data is dimensionality reduced and denoised to obtain the second matrix data, which includes: index values ​​of multiple potential influencing factors under multiple preset dimensions, and geological hazard kernel density under multiple preset dimensions;

[0210] Based on the index values ​​of multiple potential influencing factors under multiple preset dimensions and the nuclear density of geological hazards under multiple preset dimensions, multiple target influencing factors are determined from multiple potential influencing factors.

[0211] In an optional implementation, the determining module 603 is specifically used for:

[0212] Based on the index values ​​of multiple potential influencing factors under multiple preset dimensions, construct the factor vector of each potential influencing factor;

[0213] A kernel density spatial vector is constructed based on the kernel density of geological hazards under multiple preset dimensions.

[0214] Calculate the similarity between the factor vector and the kernel density space vector;

[0215] Based on similarity, multiple target impact factors are identified from multiple potential impact factors.

[0216] In an optional implementation, the determining module 603 is specifically used for:

[0217] Based on similarity, multiple potential impact factors are classified into grades, resulting in a classification result for each potential impact factor. The classification result is used to indicate the importance of each potential impact factor.

[0218] Based on the classification results, multiple target impact factors are identified from a number of potential impact factors.

[0219] In an alternative implementation, the determining module 603 is further configured to:

[0220] Identify the positive and negative influencing factors of geological hazards from multiple potential influencing factors;

[0221] The device also includes:

[0222] The processing module 604 is used to process the index values ​​under the positive influence factor and the index values ​​under the negative influence factor using the first range method and the second range method respectively, so as to obtain the standardized index values ​​of the positive influence factor and the standardized index values ​​of the negative influence factor.

[0223] Module 603 is specifically used for:

[0224] The first matrix data is constructed based on the standardized index values ​​of positive and negative impact factors and the geological hazard kernel density of multiple target slopes.

[0225] In an alternative implementation, the determining module 603 is further configured to:

[0226] Correlation analysis was conducted on the index values ​​under multiple potential influencing factors and the geological hazard kernel density of multiple target slopes to obtain the analysis results;

[0227] If the analysis results meet the preset conditions, then principal component analysis is used to reduce the dimensionality and denoise the first matrix data to obtain the second matrix data.

[0228] The processing flow of each module in the device and the interaction flow between each module can be referred to the relevant descriptions in the above method embodiments, and will not be detailed here.

[0229] Figure 9 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 9 As shown, it includes: a processor 701, a memory 702 and a bus 703. The memory 702 stores machine-readable instructions that can be executed by the processor 701. When the electronic device is running, the processor 701 communicates with the memory 702 through the bus 703. The processor 701 executes the machine-readable instructions to perform the above-mentioned risk prediction method.

[0230] This application also provides a computer-readable storage medium storing a computer program that is executed by a processor, wherein the processor performs the aforementioned risk prediction method.

[0231] In this embodiment, the computer program, when run by the processor, can also execute other machine-readable instructions to perform other methods as described in the embodiments. For details on the specific execution steps and principles, please refer to the description of the embodiments, which will not be repeated here.

[0232] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0233] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0234] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0235] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0236] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0237] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A risk prediction method, characterized in that, include: Obtain the index values ​​of the slope to be monitored under multiple target influencing factors for preset geological hazards; The weighted radar chart method is used to predict the risk of the preset geological disasters occurring on the slope to be monitored, based on the preset weights of the multiple target influence factors and the index values ​​of the slope to be monitored under the multiple target influence factors. Before predicting the risk of the preset geological hazard on the slope under monitoring using the weighted radar chart method, based on the preset weights of the multiple target influence factors and the index values ​​of the slope under the multiple target influence factors, the method further includes: Obtain the subjective and objective influence weights of the multiple target influencing factors; Based on the subjective influence weight and the objective influence weight, the preset weights of the multiple target influence factors are obtained; Before obtaining the subjective influence weights and objective influence weights of the plurality of target influence factors, the method further includes: Identify multiple potential influencing factors for the occurrence of the preset geological disaster in the area where the slope to be monitored is located; The index values ​​of multiple target slopes in the area where the pre-set geological hazard has occurred are obtained under multiple potential influencing factors, and the geological hazard kernel density of the multiple target slopes are obtained. The geological hazard kernel density of each target slope is the geological hazard kernel density of a local area including each target slope in the pre-divided area. Based on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes, a first matrix data corresponding to the multiple target slopes is constructed; The first matrix data is dimensionality reduced and denoised to obtain the second matrix data, which includes: index values ​​of the multiple potential influencing factors under multiple preset dimensions, and the geological hazard kernel density under the multiple preset dimensions; Based on the index values ​​of the multiple potential influencing factors under the multiple preset dimensions and the geological hazard kernel density under the multiple preset dimensions, the multiple target influencing factors are determined from the multiple potential influencing factors.

2. The method according to claim 1, characterized in that, The step of determining the multiple target impact factors from the multiple potential impact factors based on the index values ​​of the multiple potential impact factors under the multiple preset dimensions and the geological hazard kernel density under the multiple preset dimensions includes: Based on the index values ​​of the multiple potential influencing factors under the multiple preset dimensions, construct the factor vector of each potential influencing factor; Based on the geological hazard kernel density under the multiple preset dimensions, a kernel density space vector is constructed; Calculate the similarity between the factor vector and the kernel density space vector; Based on the similarity, the target influence factors are determined from the plurality of potential influence factors.

3. The method according to claim 2, characterized in that, The step of determining the plurality of target impact factors from the plurality of potential impact factors based on the similarity includes: Based on the similarity, the multiple potential influencing factors are classified to obtain the classification results of the multiple potential influencing factors, and the classification results are used to indicate the importance of each potential influencing factor; Based on the classification results, the target impact factors are determined from the plurality of potential impact factors.

4. The method according to claim 1, characterized in that, Before constructing the first matrix data corresponding to the multiple target slopes based on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes, the method further includes: The positive and negative influencing factors of the geological hazard are determined from the plurality of potential influencing factors; The first range method and the second range method are used to process the index values ​​under the positive influence factor and the index values ​​under the negative influence factor, respectively, to obtain the standardized index values ​​of the positive influence factor and the standardized index values ​​of the negative influence factor. The first matrix data corresponding to the multiple target slopes is constructed based on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes, including: The first matrix data is constructed based on the standardized index values ​​of the positive influence factors, the standardized index values ​​of the negative influence factors, and the geological hazard kernel density of the multiple target slopes.

5. The method according to claim 1, characterized in that, The step of reducing the dimensionality and denoising the first matrix data to obtain the second matrix data includes: A correlation analysis was conducted on the index values ​​under the multiple potential influencing factors and the geological hazard kernel density of the multiple target slopes to obtain the analysis results; If the analysis results meet the preset conditions, then principal component analysis is used to reduce the dimensionality and denoise the first matrix data to obtain the second matrix data.

6. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the risk prediction method according to any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the risk prediction method according to any one of claims 1 to 5.

Images