Landslide susceptibility assessment method and device based on secondary factor interval optimization division

Abstract

The application is suitable for the technical field of data processing, and provides a landslide sensitivity evaluation method and equipment based on secondary factor interval optimization division. The evaluation method comprises the following steps: coupling historical landslide disaster area data with continuous variable disaster-causing factors and discrete variable disaster-causing factors to obtain a continuous variable disaster-causing factor layer and a secondary factor division layer of the discrete variable disaster-causing factors; dividing the continuous variable disaster-causing factor layer to obtain a secondary factor and a secondary factor division layer; calculating the correlation between the continuous variable disaster-causing factors and the landslide disaster based on the secondary factor; calculating the correlation between the discrete variable disaster-causing factors and the landslide disaster; determining a target disaster-causing factor according to the correlation; obtaining a landslide sensitivity evaluation model based on the secondary factor division layer of the target disaster-causing factor; and evaluating by using the landslide sensitivity evaluation model to obtain a landslide sensitivity evaluation result. The landslide sensitivity evaluation method can improve the accuracy of landslide sensitivity evaluation.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a landslide sensitivity assessment method and device based on secondary factor interval optimization. Background Technology

[0002] Currently, with the progress of social development, weather prediction and assessment technologies have gradually matured, facilitating people's lives. However, technologies for predicting specific weather disasters are still not mature enough. Landslides, in particular, cause immense damage to human health, buildings, and topography. Current research literature primarily focuses on optimizing and improving landslide sensitivity assessment models to enhance accuracy. However, few studies have explored improving accuracy through optimizing factor processing procedures. In fact, the division of secondary factor intervals for disaster-causing factors is a crucial step in factor processing. Current methods for dividing these intervals utilize empirical methods or standard classification methods from Geographic Information Systems (GIS), such as the natural breakpoint method and the equal-interval statistical method. However, these methods only consider the data distribution of disaster-causing factors themselves, neglecting the correlation between disaster-causing factors and landslide disaster distribution, leading to poor accuracy in landslide sensitivity assessment. Summary of the Invention

[0003] This application provides a landslide sensitivity assessment method and device based on secondary factor interval optimization, which can solve the problem of poor accuracy in landslide sensitivity assessment.

[0004] In a first aspect, embodiments of this application provide a landslide sensitivity assessment method based on secondary factor interval optimization, the landslide sensitivity assessment method comprising:

[0005] By using historical landslide disaster area data of the target area, multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors of the target area are coupled to obtain multiple continuous variable disaster-causing factor layers and multiple discrete variable disaster-causing factor secondary factor division layers;

[0006] For each continuous variable disaster factor, the continuous variable disaster factor layer is divided to obtain the corresponding secondary factor and the secondary factor division layer of the continuous variable disaster factor.

[0007] For each continuous variable disaster-causing factor, the correlation between the continuous variable disaster-causing factor and the landslide disaster is calculated based on the corresponding secondary factors.

[0008] Calculate the correlation between each discrete variable causative factor and landslide disaster;

[0009] Based on the calculated correlations, the target disaster-causing factors for the target area are determined from multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors.

[0010] Based on the secondary factor division layers corresponding to the target disaster-causing factors and the historical landslide disaster area data, the random forest model is trained to obtain the landslide sensitivity assessment model.

[0011] The landslide sensitivity assessment model was used to assess the landslide sensitivity of the target area, and the landslide sensitivity assessment results of the target area were obtained.

[0012] Optionally, the continuous variable catastrophic factor layer is divided to obtain the corresponding secondary factors, and a secondary factor division layer of the continuous variable catastrophic factor is obtained, including:

[0013] Initialize the sparrow population X as follows:

[0014]

[0015] Among them, X 1，1 Let X represent the value of the catastrophic factor, a continuous variable in the first dimension, for the first sparrow. n，1 Let X represent the value of the catastrophic factor, a continuous variable for the nth sparrow in the first dimension. 1，dim Let X represent the value of the catastrophic factor, a continuous variable, in the dim dimension for the first sparrow. n，dim This represents the value of the catastrophic factor, a continuous variable, in the dim dimension for the nth sparrow.

[0016] Update the position of the i-th sparrow using the formula:

[0017]

[0018] Update the discoverer's location;

[0019] in, This represents the position of the i-th sparrow in the j-th dimension at the (t+1)-th iteration. Let represent the position of the i-th sparrow in the j-th dimension at the t-th iteration, where i = 1, 2, ..., n, n represents the total number of sparrows in the sparrow population, and j = 1, 2, ..., dim, dim represents the total number of dimensions. (iter...) max R2 represents the maximum number of iterations, α is a random number, α∈(0,1], R2 represents the warning value, ST represents the safety value, Q represents a random number, and L represents the parameter matrix.

[0020] Through the formula:

[0021]

[0022] Update the location of the newcomer;

[0023] Among them, X j，worst This represents the position of the global worst-case scenario in the j-th dimension. A represents the optimal position of the discoverer in the j-th dimension at the (t+1)-th iteration. + Represents the parameter matrix;

[0024] Through the formula:

[0025]

[0026] Update the guards' positions;

[0027] Among them, X j，best f represents the global optimal position in the j-th dimension, β represents the step control parameter, and f i f represents the current fitness. g Let f represent the global optimal fitness, where K is a random number, K∈[-1, 1], and f w This represents the worst global fitness, where ε is a constant.

[0028] The position of the i-th sparrow in the j-th dimension at the t-th iteration. The interval between the j-th and (j+1)-th secondary factors of the i-th group of secondary factors of the continuous variable catastrophic factor d;

[0029] Through the formula:

[0030]

[0031] Calculate the normalized landslide density value of the region corresponding to the j-th secondary factor in the continuous variable disaster-causing factor d layer.

[0032] in, This represents the area of ​​the region corresponding to the j-th second-order factor of the continuous variable catastrophic factor d. Let d represent the historical landslide area in the region corresponding to the j-th secondary factor of the continuous variable disaster-causing factor d, where d = 1, 2, ..., D, and D represents the total number of continuous variable disaster-causing factors.

[0033] Through the objective function:

[0034]

[0035] Calculate the entropy value of the i-th group of second-order factors of the continuous variable catastrophic factor d.

[0036] The entropy value of the i-th group of second-order factors of the continuous variable catastrophic factor d. Fitness of the i-th sparrow Get the fitness F of all sparrows d :

[0037]

[0038] in, This indicates the fitness level of the first sparrow. This represents the fitness of the nth sparrow;

[0039] Determine if the maximum number of iterations has been reached. If not, then set F... d The fitness with the smallest value is taken as the global optimal fitness, and the position of the sparrow corresponding to the smallest fitness value is taken as the global best position. F... d The fitness with the largest value is taken as the global worst fitness. The position of the sparrow corresponding to the fitness with the largest value is taken as the global worst position. The global best fitness, global worst fitness, global optimal position and global worst position are updated, and the step of updating the position of the i-th sparrow is returned.

[0040] If the maximum number of iterations is reached, stop iterating and set X as the value at which the maximum number of iterations is reached. j，best The interval value between the j-th and (j+1)-th secondary factors of the continuous variable catastrophic factor d;

[0041] Based on the secondary factors of the continuous variable disaster-causing factors, the continuous variable disaster-causing factor layer is divided to obtain the secondary factor division layer of the continuous variable disaster-causing factors.

[0042] Optionally, the correlation between the continuous variable disaster-causing factors and landslide disasters can be calculated based on the second-level factors corresponding to the continuous variable disaster-causing factors, including:

[0043] Through the formula:

[0044]

[0045] Calculate the correlation between the continuous variable disaster-causing factor d and landslide disaster.

[0046] Where, x dj T represents the actual observed landslide area in the region corresponding to the j-th second-order factor of the continuous variable disaster-causing factor d. dj Let d represent the theoretical estimate of the landslide area in the region corresponding to the j-th secondary factor of the continuous variable disaster-causing factor d, where d = 1, 2, ..., D, and D represents the total number of continuous variable disaster-causing factors.

[0047] Optionally, the correlation between each discrete variable causative factor and landslide disaster can be calculated separately, including:

[0048] Through the formula:

[0049]

[0050] Calculate the correlation between the discrete variable disaster-causing factor f and landslide disaster.

[0051] Where, x fl T represents the actual observed landslide area in the region corresponding to the l-th second-order factor of the discrete variable disaster-causing factor f. fl Let f = 1, 2, ..., F, where F represents the total number of discrete disaster-causing factors, and l = 1, 2, ..., L, where L represents the number of secondary factors of the discrete disaster-causing factors.

[0052] Optionally, the target disaster-causing factor for the target area can be determined from multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors based on the calculated correlation, including:

[0053] The continuous variable disaster-causing factors and discrete variable disaster-causing factors with a correlation greater than or equal to the preset correlation threshold are used as the target disaster-causing factors for the target area.

[0054] Optionally, a random forest model is trained based on the secondary factor division layers corresponding to the target disaster-causing factor and historical landslide disaster area data to obtain a landslide sensitivity assessment model, including:

[0055] The secondary factor division layer corresponding to the target disaster-causing factor is coupled with the historical landslide disaster area data to obtain the landslide disaster layer;

[0056] A random forest model was trained using a landslide hazard layer, and the trained random forest model was used as a landslide sensitivity assessment model.

[0057] Secondly, embodiments of this application provide a landslide sensitivity assessment device based on secondary factor interval optimization, comprising:

[0058] The coupling module is used to couple multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors in the target area using historical landslide disaster area data, to obtain a layer of multiple continuous variable disaster-causing factors and a secondary factor division layer of multiple discrete variable disaster-causing factors.

[0059] The partitioning module is used to partition the continuous variable disaster factor layer for each continuous variable disaster factor, obtain the secondary factors corresponding to the continuous variable disaster factor, and obtain the secondary factor partitioning layer of the continuous variable disaster factor.

[0060] The first calculation module is used to calculate the correlation between the continuous variable disaster-causing factor and the landslide disaster based on the secondary factors corresponding to the continuous variable disaster-causing factor for each continuous variable disaster-causing factor.

[0061] The second calculation module is used to calculate the correlation between each discrete variable disaster-causing factor and landslide disaster;

[0062] The determination module identifies the target disaster-causing factor for the target area from multiple continuous and multiple discrete disaster-causing factors based on the calculated correlation.

[0063] The training module is used to train the random forest model based on the secondary factor division layers corresponding to the target disaster-causing factors and historical landslide disaster area data to obtain the landslide sensitivity assessment model.

[0064] The assessment module uses a landslide sensitivity assessment model to assess the landslide sensitivity of the target area and obtain the landslide sensitivity assessment results for the target area.

[0065] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the landslide sensitivity assessment method based on secondary factor interval optimization.

[0066] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned landslide sensitivity assessment method based on secondary factor interval optimization.

[0067] The above-mentioned solution in this application has the following beneficial effects:

[0068] In the embodiments of this application, multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors in the target area are coupled using historical landslide disaster area data to obtain multiple continuous variable disaster-causing factor layers and multiple discrete variable disaster-causing factor secondary factor partitioning layers. Then, for each continuous variable disaster-causing factor, the continuous variable disaster-causing factor layer is partitioned to obtain the corresponding secondary factor and the secondary factor partitioning layer. Then, for each continuous variable disaster-causing factor, the correlation between the continuous variable disaster-causing factor and the landslide disaster is calculated based on the corresponding secondary factor. The correlation between each discrete variable disaster-causing factor and the landslide disaster is also calculated. Then, based on the calculated correlation, the target disaster-causing factor in the target area is determined from the multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors. Then, a random forest model is trained based on the secondary factor partitioning layer corresponding to the target disaster-causing factor and historical landslide disaster area data to obtain a landslide sensitivity assessment model. Finally, the landslide sensitivity assessment model is used to assess the landslide sensitivity of the target area to obtain the landslide sensitivity assessment result of the target area. In this process, each continuous variable disaster-causing factor layer is divided to obtain the corresponding secondary factors, which can improve the importance of the obtained secondary factors. Multiple target disaster-causing factors are obtained based on correlation, and target disaster-causing factors with high correlation to landslide disasters can be screened out. This results in high assessment accuracy of the landslide sensitivity assessment model trained based on the target disaster-causing factors, and thus greatly improves the accuracy of the landslide sensitivity assessment completed based on the landslide sensitivity assessment model.

[0069] Other beneficial effects of this application will be described in detail in the following detailed description section. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0071] Figure 1 A flowchart of a landslide sensitivity assessment method based on second-level factor interval optimization division provided in an embodiment of this application;

[0072] Figure 2 A continuous variable hazard factor layer provided in one embodiment of this application;

[0073] Figure 3 A landslide hazard sensitivity zoning map provided in one embodiment of this application;

[0074] Figure 4 This application provides a layer for dividing continuous variable hazard factors into secondary factors, as an embodiment of the present application.

[0075] Figure 5 A schematic diagram of a landslide sensitivity assessment device based on second-level factor interval optimization division provided in an embodiment of this application;

[0076] Figure 6 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0077] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0078] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0079] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0080] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0081] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0082] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0083] To address the issue of poor accuracy in existing landslide sensitivity assessments, this application provides a landslide sensitivity assessment method based on optimized partitioning of secondary factor intervals. This method couples multiple continuous and discrete landslide hazard-causing factors within the target area using historical landslide disaster area data, resulting in layers of continuous and discrete hazard-causing factors and secondary factor partitioning layers. Then, for each continuous hazard-causing factor, the continuous hazard-causing factor layer is further partitioned to obtain the corresponding secondary factors, thus achieving the secondary factor partitioning of the continuous hazard-causing factors. The system first divides the data into layers, then calculates the correlation between each continuous variable disaster-causing factor and landslide disaster based on its corresponding secondary factors. The correlation between each discrete variable disaster-causing factor and landslide disaster is also calculated. Based on these correlations, target disaster-causing factors for the target area are determined from multiple continuous and discrete disaster-causing factors. Then, a random forest model is trained using the secondary factors corresponding to the target disaster-causing factors and historical landslide disaster area data to obtain a landslide sensitivity assessment model. Finally, this model is used to assess the landslide sensitivity of the target area, yielding the assessment results. The process of dividing the data into layers for each continuous variable disaster-causing factor and obtaining its corresponding secondary factors increases the importance of these secondary factors. Obtaining multiple target disaster-causing factors based on correlations allows for the selection of those with high correlation to landslide disasters, thus improving the accuracy of the landslide sensitivity assessment model trained on these target disaster-causing factors. This significantly enhances the overall accuracy of the landslide sensitivity assessment performed using this model.

[0084] The landslide sensitivity assessment method provided in this application will be illustrated below.

[0085] like Figure 1 As shown, the landslide sensitivity assessment method provided in this application includes the following steps:

[0086] Step 11: Using historical landslide disaster area data of the target area, couple multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors of the target area to obtain multiple continuous variable disaster-causing factor layers and multiple discrete variable disaster-causing factor secondary factor division layers.

[0087] The aforementioned target area refers to the area requiring landslide sensitivity assessment; the aforementioned historical landslide disaster area data includes the area of ​​the target area where landslides occurred during the historical time period; the causative factor refers to the factors that influence the occurrence of the disaster; the aforementioned continuous variable causative factor refers to a causative factor that can take any value within a certain range, for example: the topographic humidity index ranges from -0.31 to 22.11, and within this range, the topographic humidity index can be any value, so the topographic humidity index is a continuous variable causative factor; the aforementioned discrete variable causative factor refers to a causative factor with fixed secondary factors, for example: the secondary factors of lithology are stable rock, evenly slidable rock, moderately slidable rock, and easily slidable rock, so lithology is a discrete variable causative factor. In some embodiments of this application, by coupling the continuous variable causative factor and the historical landslide disaster area data, a continuous variable causative factor layer corresponding to the continuous variable causative factor can be obtained. Similarly, by coupling discrete variable disaster-causing factors and historical landslide disaster area data, since the values ​​of discrete variable disaster-causing factors are discrete, their discrete values ​​are secondary factors, thus the secondary factors are fixed. Therefore, after coupling, a secondary factor partitioning layer corresponding to the discrete variable disaster-causing factor can be obtained. Here, coupling refers to the overlay operation of historical landslide disaster area data and disaster-causing factor layers using a Geographic Information System (GIS); the continuous variable disaster-causing factor layer is used to represent the relationship between continuous variable disaster-causing factors and landslide disaster area; the secondary factor partitioning layer of discrete variable disaster-causing factors is used to represent the relationship between discrete variable disaster-causing factors and landslide disaster area.

[0088] For example, when the target area is Panzhihua City, Sichuan Province, ten influencing factors can be obtained, including lithology, elevation, slope, surface relief, topography, topographic humidity index, normalized difference vegetation index, annual average rainfall, distance from water systems, and distance from roads, as well as historical landslide disaster area data for Panzhihua City. Among these ten disaster-causing factors, six factors—elevation, slope, surface relief, topographic humidity index, normalized difference vegetation index, and annual average rainfall—are continuous variables, while the other four are discrete variables. The values ​​for distance from water systems and distance from roads need to be determined based on the actual conditions of the target area. Each disaster-causing factor is coupled with the historical landslide disaster area data to obtain a layer of six continuous variable disaster-causing factors and a secondary factor layer of four discrete variable disaster-causing factors.

[0089] The following example illustrates the aforementioned continuous variable disaster factor layer.

[0090] Continuous variable hazard factor layer such as Figure 2 As shown in the figure, the continuous variable disaster-causing factors layers are displayed, with elevation values ​​ranging from 910 meters (m) to 3971 meters (high), slope values ​​ranging from 0 degrees (°) to 56.63° (high), surface relief values ​​ranging from 0 meters (high) to 679.74 meters (high), topographic humidity index values ​​ranging from -0.31 to 22.11 (high), normalized difference vegetation index values ​​ranging from 0.22 to 0.90 (high), and annual average rainfall values ​​ranging from 821 millimeters (mm) to 1147 millimeters (high). The area above the layers is north (N), and the scale is 1 cm: 10 km.

[0091] It is worth mentioning that the coupled continuous variable disaster-causing factor layer and the secondary factor division layer of discrete variable disaster-causing factors can intuitively and accurately represent the relationship between disaster-causing factors and landslide disaster area.

[0092] Step 12: For each continuous variable disaster factor, divide the continuous variable disaster factor layer to obtain the corresponding secondary factor and the secondary factor division layer of the continuous variable disaster factor.

[0093] In some embodiments of this application, the sparrow search algorithm can be used to obtain the secondary factors corresponding to the continuous variable disaster factors, and the continuous variable disaster factor layer can be divided according to the secondary factors of the continuous variable disaster factors to obtain the secondary factor division layer of the continuous variable disaster factors.

[0094] It is worth mentioning that the numerical characteristics of continuous variable disaster-causing factors are that they can take any value within the range of values. Therefore, they cannot be directly used to calculate the correlation between continuous variable disaster-causing factors and landslide disasters. The secondary factors corresponding to the continuous variable disaster-causing factors obtained by using the sparrow algorithm solve the problem that they cannot be directly used to calculate the correlation between continuous variable disaster-causing factors and landslide disasters, and the importance of the obtained secondary factors is improved.

[0095] Step 13: For each continuous variable disaster-causing factor, calculate the correlation between the continuous variable disaster-causing factor and the landslide disaster based on the corresponding secondary factors.

[0096] In some embodiments of this application, the chi-square test can be used to calculate the correlation between continuous variable disaster-causing factors and landslide disasters.

[0097] Specifically, through the formula:

[0098]

[0099] Calculate the correlation between the continuous variable disaster-causing factor d and landslide disaster.

[0100] Where, x dj T represents the actual observed landslide area in the region corresponding to the j-th second-order factor of the continuous variable disaster-causing factor d. dj Let d represent the theoretical estimate of the landslide area in the region corresponding to the j-th secondary factor of the continuous variable disaster-causing factor d, where d = 1, 2, ..., D, and D represents the total number of continuous variable disaster-causing factors.

[0101] It should be noted that the above actual observed values ​​are the landslide areas in the regions corresponding to the j-th secondary factor of the continuous variable disaster-causing factor d, obtained directly from the secondary factor division layer of the continuous variable disaster-causing factor d. The above theoretically predicted values ​​are the landslide areas in the regions corresponding to the j-th secondary factor of the continuous variable disaster-causing factor d, assuming that the continuous variable disaster-causing factor d is not correlated with landslide disasters. The larger the value, the less correct the assumption that the continuous variable disaster-causing factor d is not related to landslide disasters; that is, the greater the correlation between the continuous variable disaster-causing factor d and landslide disasters.

[0102] Specifically, the above theoretical prediction is calculated as follows: assuming that the continuous variable disaster-causing factor d is not related to the landslide disaster, the landslide area of ​​the target area is divided by the total area of ​​the target area to obtain the landslide density. Then, the area of ​​the region corresponding to the j-th secondary factor of the continuous variable disaster-causing factor d is multiplied by the landslide density to obtain the theoretical prediction value of the landslide area in the region corresponding to the j-th secondary factor of the continuous variable disaster-causing factor d.

[0103] It is worth mentioning that using the chi-square test to calculate the correlation between continuous variable disaster-causing factors and landslide disasters can improve the accuracy of the correlation.

[0104] Step 14: Calculate the correlation between each discrete variable disaster-causing factor and landslide disaster.

[0105] In some embodiments of this application, the chi-square test can be used to calculate the correlation between discrete variable disaster-causing factors and landslide disasters.

[0106] Specifically, through the formula:

[0107]

[0108] Calculate the correlation between the discrete variable disaster-causing factor f and landslide disaster.

[0109] Where, x fl T represents the actual observed landslide area in the region corresponding to the l-th second-order factor of the discrete variable disaster-causing factor f. fl Let f = 1, 2, ..., F, where F represents the total number of discrete disaster-causing factors, and l = 1, 2, ..., L, where L represents the number of secondary factors of the discrete disaster-causing factors.

[0110] It should be noted that the above actual observed values ​​are: the landslide area in the region corresponding to the l-th second-level factor of the discrete variable disaster-causing factor f, directly obtained from the second-level factor partitioning layer corresponding to the discrete variable disaster-causing factor f. The above theoretically predicted values ​​are: the landslide area in the region corresponding to the l-th second-level factor of the discrete variable disaster-causing factor f, assuming that the discrete variable disaster-causing factor f is unrelated to landslide disasters. The larger the value, the less correct the assumption that the discrete variable disaster-causing factor f is not related to landslide disasters; that is, the greater the correlation between the discrete variable disaster-causing factor f and landslide disasters.

[0111] It is worth mentioning that using the chi-square test to calculate the correlation between discrete variable disaster-causing factors and landslide disasters can improve the accuracy of the correlation.

[0112] Step 15: Determine the target disaster-causing factor for the target area from multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors based on the calculated correlation.

[0113] In some embodiments of this application, continuous variable disaster-causing factors and discrete variable disaster-causing factors with a correlation greater than or equal to a preset correlation threshold are used as target disaster-causing factors for the target area.

[0114] For example, if the preset correlation threshold value for the continuous variable disaster-causing factor d is y, when the correlation between the continuous variable disaster-causing factor d and the landslide disaster... When the value is greater than or equal to y, the continuous variable disaster-causing factor d is the target disaster-causing factor; if the preset correlation threshold value for the discrete variable disaster-causing factor f is z, then the correlation between the discrete variable disaster-causing factor f and the landslide disaster... When z is greater than or equal to z, the discrete variable catastrophic factor f is the target catastrophic factor.

[0115] It is worth mentioning that determining the target disaster-causing factors in the target area based on correlation can retain disaster-causing factors with a high degree of correlation with landslide disasters and eliminate disaster-causing factors with a low degree of correlation, thereby improving the accuracy of the data.

[0116] Step 16: Train the random forest model based on the secondary factor division layer corresponding to the target disaster-causing factor and the historical landslide disaster area data to obtain the landslide sensitivity assessment model.

[0117] In some embodiments of this application, the secondary factor partitioning layers corresponding to the target disaster-causing factors (the secondary factor partitioning layers of continuous variable disaster-causing factors and the secondary factor partitioning layers of discrete variable disaster-causing factors) can be coupled with historical landslide disaster area data (that is, the secondary factor partitioning layers corresponding to all target disaster-causing factors and the historical landslide disaster area data are overlaid through a geographic information system) to obtain a landslide disaster layer that represents the spatial relationship between secondary factors and landslide disasters. Then, a random forest model is trained using the landslide disaster layer, and the trained random forest model is used as a landslide sensitivity assessment model.

[0118] Specifically, the landslide hazard layer is used as training data for the random forest model. Random samples with replacement are randomly selected from the training data, and these selected samples are used to train a decision tree, serving as the root node of the decision tree. The decision tree represents the relationship between the sample value and the sample's features. In the landslide hazard layer, the hazard-causing factor is the sample's feature, and whether a landslide has occurred is the sample's value. A subset of features is randomly selected from the sample's features, and a Gini coefficient or similar strategy is used to select one feature as the splitting feature for that node. Each node is split according to this process until it can no longer be split. This process is repeated to build the decision tree. When the number of decision trees in the random forest model reaches a preset number, training stops, resulting in the landslide sensitivity assessment model.

[0119] It is worth mentioning that training a random forest model using a landslide hazard layer can improve the accuracy of landslide sensitivity assessment models.

[0120] Step 17: Use the landslide sensitivity assessment model to assess the landslide sensitivity of the target area and obtain the landslide sensitivity assessment results for the target area.

[0121] In some embodiments of this application, a landslide sensitivity assessment model can be used to predict the probability of a landslide disaster occurring in each unit area of ​​the target area, obtaining the unit landslide probability. The value range of the unit landslide probability is [0,1], and the larger the value, the higher the probability of a landslide. Then, all unit landslide probabilities are integrated onto a single graph to obtain the landslide sensitivity assessment result for the target area. Specifically, during the prediction process, the values ​​of the disaster-causing factors (i.e., the aforementioned target disaster-causing factors) of the unit areas of the target area can be input into the landslide sensitivity assessment model to obtain the unit landslide probability corresponding to that unit area.

[0122] The aforementioned unit region refers to dividing the target area into multiple regions, with each region constituting a unit region. For example, a unit region might have disaster-causing factors such as an elevation of 1200m, a surface relief of 68m, and a normalized vegetation index of 0.76. These disaster-causing factors are input into a landslide sensitivity assessment model to obtain the corresponding unit landslide probability. This calculation is performed for each unit region, and finally, all unit landslide probabilities are integrated onto a single map to obtain the landslide sensitivity assessment result for the target area. Based on a preset range, the unit landslide probabilities are classified into four levels: extremely low sensitivity, low sensitivity, moderate sensitivity, and high sensitivity. For instance, if a unit region has a unit landslide probability of 0.13, and the preset range for extremely low sensitivity is 0 to 0.2, then the unit landslide probability is classified as extremely low sensitivity.

[0123] The following example illustrates the landslide sensitivity assessment results.

[0124] The landslide hazard sensitivity zoning map provided in one embodiment of this application (i.e., the landslide sensitivity assessment result mentioned above) is as follows: Figure 3 As shown, the landslide sensitivity level of the target area is divided into four levels: extremely low sensitivity, low sensitivity, moderate sensitivity, and high sensitivity. By coupling historical landslide disaster area data of the target area with the landslide disaster sensitivity zoning map, and using the Statistical Product and Service Solutions (SPSS) platform to plot the receiver operating characteristic (ROC) curve, the area under the curve was found to be 0.804, indicating that the landslide disaster sensitivity zoning map has high reliability.

[0125] It is worth mentioning that dividing each continuous variable disaster-causing factor layer into secondary factors can increase the importance of the secondary factors. By obtaining multiple target disaster-causing factors based on correlation, it is possible to screen out target disaster-causing factors that are highly correlated with landslide disasters. This results in high accuracy of the landslide sensitivity assessment model trained based on the target disaster-causing factors, and thus greatly improves the accuracy of the landslide sensitivity assessment completed based on the landslide sensitivity assessment model.

[0126] The above landslide sensitivity assessment results can be used to predict areas in the target region where landslides may occur in a timely manner, prevent the risk of landslides, and reduce the losses caused by landslides.

[0127] The specific steps of step 12 described above will be illustrated below with reference to specific embodiments.

[0128] In some embodiments of this application, the specific implementation process of step 12 above includes the following steps:

[0129] Step 12.1: Use the sparrow search algorithm to obtain the secondary factors corresponding to the disaster-causing factors of the continuous variables.

[0130] The first step is to initialize the sparrow population X as follows:

[0131]

[0132] Among them, X 1，1 Let X represent the value of the catastrophic factor, a continuous variable in the first dimension, for the first sparrow. n，1 Let X represent the value of the catastrophic factor, a continuous variable for the nth sparrow in the first dimension. 1，dim Let X represent the value of the catastrophic factor, a continuous variable, in the dim dimension for the first sparrow. n，dim This represents the value of the catastrophic factor, a continuous variable, in the dim dimension for the nth sparrow.

[0133] The second step is to update the position of the i-th sparrow using the formula:

[0134]

[0135] Update the discoverer's location;

[0136] in, This represents the position of the i-th sparrow in the j-th dimension at the (t+1)-th iteration. Let represent the position of the i-th sparrow in the j-th dimension at the t-th iteration, where i = 1, 2, ..., n, n represents the total number of sparrows in the sparrow population, and j = 1, 2, ..., dim, dim represents the total number of dimensions. (iter...) max R represents the maximum number of iterations, α is a random number, α∈(0,1], R2 represents the warning value, ST represents the safety value, Q represents a random number that follows a normal distribution, and L represents a parameter matrix in which each element is 1;

[0137] Through the formula:

[0138]

[0139] Update the location of the newcomer;

[0140] Among them, X j，worst This represents the position of the global worst-case scenario in the j-th dimension. A represents the optimal position of the discoverer in the j-th dimension at the (t+1)-th iteration. + Let A represent the parameter matrix. + =A T (AA T ) -1 A represents a parameter matrix with elements of 1 or -1. T This represents the transpose operation on matrix A;

[0141] Through the formula:

[0142]

[0143] Update the guards' positions;

[0144] Among them, X j，best Let f represent the global optimal position in the j-th dimension, β represent the step control parameter, and f is a random number following a normal distribution. i f represents the current fitness. g Let f represent the global optimal fitness, where K is a random number, K∈[-1, 1], and f w This represents the worst global fitness, where ε is a constant.

[0145] The third step is to determine the position of the i-th sparrow in the j-th dimension at the t-th iteration. The interval between the j-th and (j+1)-th secondary factors of the i-th group of secondary factors of the continuous variable catastrophic factor d;

[0146] Through the formula:

[0147]

[0148] Calculate the normalized landslide density value of the region corresponding to the j-th secondary factor in the continuous variable disaster-causing factor d layer.

[0149] in, This represents the area of ​​the region corresponding to the j-th second-order factor of the continuous variable catastrophic factor d. Let d represent the historical landslide area in the region corresponding to the j-th secondary factor of the continuous variable disaster-causing factor d, where d = 1, 2, ..., D, and D represents the total number of continuous variable disaster-causing factors.

[0150] Through the objective function:

[0151]

[0152] Calculate the entropy value of the i-th group of second-order factors of the continuous variable catastrophic factor d.

[0153] The entropy value of the i-th group of second-order factors of the continuous variable catastrophic factor d. Fitness of the i-th sparrow Get the fitness F of all sparrows d :

[0154]

[0155] in, This indicates the fitness level of the first sparrow. This represents the fitness of the nth sparrow.

[0156] The fourth step is to determine whether the maximum number of iterations has been reached. If the maximum number of iterations has not been reached, then F... d The fitness with the smallest value is taken as the global optimal fitness, and the position of the sparrow corresponding to the smallest fitness value is taken as the global best position. F... d The fitness with the largest value is taken as the global worst fitness. The position of the sparrow corresponding to the fitness with the largest value is taken as the global worst position. The global best fitness, global worst fitness, global optimal position and global worst position are updated, and the step of updating the position of the i-th sparrow is returned.

[0157] If the maximum number of iterations is reached, stop iterating and set X as the value at which the maximum number of iterations is reached. j，bestThe interval value between the j-th and (j+1)-th secondary factors of the continuous variable hazard factor d.

[0158] For example, the Sparrow Algorithm above will eventually output three values: 10, 30, and 50. The four secondary factors are: less than 10, 10 to 30, 30 to 50, and greater than 50.

[0159] It should be noted that in some embodiments of this application, entropy calculation can be used as the objective function. The smaller the entropy, the greater the dispersion of the disaster-causing factor, and the greater the importance of the corresponding secondary factor. Therefore, in the process of the sparrow search algorithm, the entropy becomes smaller and smaller, and finally the secondary factor with the greatest importance is obtained.

[0160] Step 12.2: Based on the secondary factors of the continuous variable disaster-causing factors, divide the continuous variable disaster-causing factor layer to obtain the secondary factor division layer of the continuous variable disaster-causing factors.

[0161] Specifically, the secondary factors obtained from the above steps divide the continuous variable disaster-causing factors into multiple intervals. Based on the divided intervals, the area range of the continuous variable disaster-causing factor layer is re-divided to obtain the secondary factor division layer of the continuous variable disaster-causing factors.

[0162] For example, the secondary factors of elevation are less than 1522m, 1522m to 1990m, 1990m to 2542m, and greater than 2542m. Based on the secondary factors, the area range of the continuous variable disaster factor layer of elevation is re-divided to obtain the secondary factor division layer of elevation.

[0163] The following example illustrates the secondary factor division layer of the aforementioned continuous variable disaster-causing factors.

[0164] An embodiment of this application provides a secondary factor partitioning layer for continuous variable hazard factors, as shown below. Figure 4As shown, the secondary factors for elevation are less than 1522m, 1522m to 1990m, 1990m to 2542m, and greater than 2542m; the secondary factors for slope are less than 9.82°, 9.82° to 17.36°, 17.36° to 25.57°, and greater than 25.57°; the secondary factors for surface relief are less than 59m, 59m to 107m, 107m to 165m, and greater than 165m. The secondary factors of the topographic humidity index are less than 4.87, 4.87 to 9.18, 9.18 to 12.43, and greater than 12.43; the secondary factors of the normalized difference vegetation index are less than 0.62, 0.62 to 0.75, 0.75 to 0.84, and greater than 0.84; the average annual rainfall is less than 951 mm, 951 mm to 1030 mm, 1030 mm to 1082 mm, and greater than 1082 mm. Above this layer is North (N), with a scale of 1 cm: 10 km.

[0165] It can be seen that obtaining secondary factors based on the sparrow search algorithm can improve the importance of secondary factors, and the secondary factor partitioning layer of continuous variable disaster factors obtained based on secondary factor partitioning has high accuracy.

[0166] The following is an exemplary description of a landslide sensitivity assessment device based on secondary factor interval optimization division provided in this application.

[0167] like Figure 5 As shown, this application provides a landslide sensitivity assessment device based on secondary factor interval optimization partitioning. The landslide sensitivity assessment device 500 includes:

[0168] The coupling module 501 is used to couple multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors in the target area using historical landslide disaster area data of the target area, to obtain a layer of multiple continuous variable disaster-causing factors and a secondary factor division layer of multiple discrete variable disaster-causing factors.

[0169] The partitioning module 502 is used to partition the continuous variable disaster factor layer for each continuous variable disaster factor, obtain the secondary factor corresponding to the continuous variable disaster factor, and obtain the secondary factor partitioning layer of the continuous variable disaster factor.

[0170] The first calculation module 503 is used to calculate the correlation between the continuous variable disaster-causing factor and the landslide disaster based on the secondary factors corresponding to the continuous variable disaster-causing factor for each continuous variable disaster-causing factor.

[0171] The second calculation module 504 is used to calculate the correlation between each discrete variable disaster-causing factor and landslide disaster;

[0172] Module 505 determines the target disaster-causing factor for the target area from multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors based on the calculated correlation.

[0173] Training module 506 is used to train the random forest model based on the secondary factor division layer corresponding to the target disaster-causing factor and historical landslide disaster area data to obtain the landslide sensitivity assessment model.

[0174] The assessment module 507 uses a landslide sensitivity assessment model to assess the landslide sensitivity of the target area and obtains the landslide sensitivity assessment results for the target area.

[0175] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0176] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the functions described above can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments 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. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0177] like Figure 6 As shown, an embodiment of this application provides a terminal device, wherein the terminal device D10 of this embodiment includes: at least one processor D100 ( Figure 6 The diagram shows only one processor, a memory D101, and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, wherein the processor D100 executes the computer program D102 to implement the steps in any of the above method embodiments.

[0178] Specifically, when the processor D100 executes the computer program D102, it couples multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors in the target area using historical landslide disaster area data, obtaining multiple continuous variable disaster-causing factor layers and multiple discrete variable disaster-causing factor secondary factor partitioning layers. Then, for each continuous variable disaster-causing factor, it partitions the continuous variable disaster-causing factor layer to obtain the corresponding secondary factor, and obtains the secondary factor partitioning layer for the continuous variable disaster-causing factor. Then, it further partitions the continuous variable disaster-causing factor layer for each continuous variable disaster-causing factor. This paper proposes a landslide sensitivity assessment model. Based on the secondary factors corresponding to continuous variable disaster-causing factors, the correlation between these factors and landslide disasters is calculated. The correlation between each discrete variable disaster-causing factor and landslide disasters is also calculated. Then, based on the calculated correlations, target disaster-causing factors for the target area are determined from multiple continuous and discrete disaster-causing factors. A random forest model is then trained using the secondary factors corresponding to the target disaster-causing factors and historical landslide disaster area data to obtain a landslide sensitivity assessment model. Finally, the landslide sensitivity assessment model is used to assess the landslide sensitivity of the target area, yielding the assessment results. Specifically, dividing each continuous variable disaster-causing factor into layers to obtain the corresponding secondary factors increases the importance of these secondary factors. Obtaining multiple target disaster-causing factors based on correlations allows for the selection of those with high correlation to landslide disasters, thus improving the accuracy of the landslide sensitivity assessment model trained based on these target disaster-causing factors. This significantly enhances the overall accuracy of the landslide sensitivity assessment performed using this model.

[0179] The processor D100 can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0180] In some embodiments, the memory D101 may be an internal storage unit of the terminal device D10, such as a hard disk or memory of the terminal device D10. In other embodiments, the memory D101 may be an external storage device of the terminal device D10, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device D10. Furthermore, the memory D101 may include both internal and external storage units of the terminal device D10. The memory D101 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory D101 can also be used to temporarily store data that has been output or will be output.

[0181] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0182] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0183] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a landslide sensitivity assessment method / terminal device based on second-order factor interval optimization partitioning, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0184] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0185] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0186] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A landslide sensitivity assessment method based on second-order factor interval optimization, characterized in that, include: By using historical landslide disaster area data of the target area, multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors of the target area are coupled to obtain multiple continuous variable disaster-causing factor layers and multiple discrete variable disaster-causing factor secondary factor division layers; Coupling refers to the overlay operation of historical landslide disaster area data with disaster-causing factors through a geographic information system; For each continuous variable disaster-causing factor, the continuous variable disaster-causing factor layer is divided to obtain the secondary factor corresponding to the continuous variable disaster-causing factor, and the secondary factor division layer of the continuous variable disaster-causing factor is obtained. For each continuous variable disaster-causing factor, the correlation between the continuous variable disaster-causing factor and the landslide disaster is calculated based on the secondary factors corresponding to the continuous variable disaster-causing factor. Calculate the correlation between each discrete variable causative factor and landslide disaster; Based on the calculated correlation, the target disaster-causing factor for the target area is determined from the plurality of continuous variable disaster-causing factors and the plurality of discrete variable disaster-causing factors. The random forest model is trained based on the secondary factor division layer corresponding to the target disaster-causing factor and the historical landslide disaster area data to obtain the landslide sensitivity assessment model. The landslide sensitivity assessment model is used to assess the landslide sensitivity of the target area, and the landslide sensitivity assessment results of the target area are obtained. The step of dividing the continuous variable disaster-causing factor layer to obtain the secondary factors corresponding to the continuous variable disaster-causing factors, and obtaining the secondary factor division layer of the continuous variable disaster-causing factors, includes: Initialize the sparrow population for: in, This represents the value of the catastrophic factor, a continuous variable, in the first dimension for the first sparrow. Indicates the first The sparrow only considers the values ​​of the catastrophic factors, which are continuous variables, in the first dimension. This indicates that the first sparrow was in the [number]th [location]. The values ​​of the catastrophic factors of the continuous variables mentioned in the document. Indicates the first Only sparrows in the first The values ​​of the disaster-causing factors of the continuous variables mentioned in the document; Update # The location of the sparrow is determined by the formula: Update the discoverer's location; in, Indicates the first Only sparrows in the first Wei Zhongyu Position at the next iteration Indicates the first Only sparrows in the first Wei Zhongyu Position at the next iteration , This represents the total number of sparrows in the sparrow population. , This represents the total number of dimensions. Indicates the maximum number of iterations. It is a random number. , Indicates the warning value. Indicates a safe value. Represents a random number. Represents the parameter matrix; Through the formula: Update the location of the newcomer; in, Indicates the first The worst position in the dimension. Indicates the discoverer was in the During the nth iteration, at the... The best position in the dimensional, Represents the parameter matrix; Through the formula: Update the guards' positions; in, Indicates the first The globally optimal position in the dimension. Indicates the step control parameters. Indicates the current fitness. This represents the globally optimal fitness. It is a random number. , Indicates the worst global fitness. It is a constant; The first Only sparrows in the first Wei Zhongyu Position at the next iteration As the continuous variable causative factor The Group secondary factors The second-order factor and the first The interval values ​​between the secondary factors; Through the formula: Calculate the continuous variable hazard factor In the corresponding continuous variable hazard factor layer, the first Normalized values ​​of landslide density in the regions corresponding to each secondary factor ; in, The continuous variable hazard factor The The area of ​​the region corresponding to each second-level factor The continuous variable hazard factor The The historical landslide area in the region corresponding to each secondary factor , This represents the total number of disaster-causing factors among the continuous variables; Through the objective function: Calculate the continuous variable hazard factor The Entropy of group second-order factors ; The continuous variable catastrophic factors The Entropy of group second-order factors As the first The adaptability of a sparrow Get the fitness of all sparrows. : in, This indicates the fitness level of the first sparrow. Indicates the first The adaptability of a single sparrow; Determine whether the number of iterations has reached the maximum number of iterations. If the maximum number of iterations has not been reached, then... The fitness with the smallest value is taken as the global optimal fitness, and the position of the sparrow corresponding to the smallest fitness value is taken as the global best position. The fitness with the largest median fitness value is taken as the global worst fitness value. The position of the sparrow corresponding to the fitness with the largest median fitness value is taken as the global worst position. The global optimal fitness value, the global worst fitness value, the global best position, and the global worst position are updated, and the update is returned. The steps to locate the sparrow; If the maximum number of iterations is reached, then the iteration stops, and the value at which the maximum number of iterations was reached is recorded. As the continuous variable causative factor The The second-order factor and the first The interval values ​​between the secondary factors; Based on the secondary factors of the continuous variable disaster-causing factors, the continuous variable disaster-causing factor layer is divided to obtain the secondary factor division layer of the continuous variable disaster-causing factors.

2. The landslide sensitivity assessment method according to claim 1, characterized in that, The calculation of the correlation between the continuous variable disaster-causing factor and the landslide disaster based on the secondary factors corresponding to the continuous variable disaster-causing factor includes: Through the formula: Calculate the continuous variable hazard factor Correlation with landslide disasters ; in, The continuous variable hazard factor The The actual observed values ​​of landslide area in the region corresponding to each secondary factor. The continuous variable hazard factor The The theoretical estimated value of landslide area in the region corresponding to each secondary factor. , This represents the total number of disaster-causing factors in the continuous variable.

3. The landslide sensitivity assessment method according to claim 1, characterized in that, The calculation of the correlation between each discrete variable disaster-causing factor and landslide disaster includes: Through the formula: Calculate the catastrophic factors of the discrete variables. Correlation with landslide disasters ; in, The discrete variable represents the catastrophic factor. The The actual observed values ​​of landslide area in the region corresponding to each secondary factor. Disaster-causing factors of discrete variables The The theoretical estimated value of landslide area in the region corresponding to each secondary factor. , This represents the total number of catastrophic factors among the discrete variables. , This indicates the number of second-order factors of the catastrophic factor in the discrete variable.

4. The landslide sensitivity assessment method according to claim 1, characterized in that, The step of determining the target disaster-causing factor for the target area from the plurality of continuous variable disaster-causing factors and the plurality of discrete variable disaster-causing factors based on the calculated correlation includes: The continuous variable disaster-causing factors and discrete variable disaster-causing factors with a correlation greater than or equal to the preset correlation threshold are used as the target disaster-causing factors for the target region.

5. The landslide sensitivity assessment method according to claim 1, characterized in that, The random forest model is trained based on the secondary factor partitioning layer corresponding to the target disaster-causing factor and the historical landslide disaster area data to obtain a landslide sensitivity assessment model, including: The secondary factor division layer corresponding to the target disaster-causing factor is coupled with the historical landslide disaster area data to obtain the landslide disaster layer; The random forest model is trained using the landslide hazard layer, and the trained random forest model is used as a landslide sensitivity assessment model.

6. A landslide sensitivity assessment device based on second-level factor interval optimization partitioning, characterized in that, include: The coupling module is used to couple multiple continuous variable disaster-causing factors and multiple discrete variable disaster-causing factors in the target area using historical landslide disaster area data, respectively, to obtain a layer of multiple continuous variable disaster-causing factors and a secondary factor division layer of multiple discrete variable disaster-causing factors; Coupling refers to the overlay operation of historical landslide disaster area data with disaster-causing factors through a geographic information system; The partitioning module is used to partition the continuous variable disaster-causing factor layer for each continuous variable disaster-causing factor to obtain the secondary factor corresponding to the continuous variable disaster-causing factor, and to obtain the secondary factor partitioning layer of the continuous variable disaster-causing factor. The first calculation module is used to calculate the correlation between the continuous variable disaster-causing factor and the landslide disaster based on the secondary factor corresponding to the continuous variable disaster-causing factor for each continuous variable disaster-causing factor. The second calculation module is used to calculate the correlation between each discrete variable disaster-causing factor and landslide disaster; The determination module determines the target disaster-causing factor of the target area from the multiple continuous variable disaster-causing factors and the multiple discrete variable disaster-causing factors based on the calculated correlation. The training module is used to train the random forest model based on the secondary factor division layer corresponding to the target disaster-causing factor and the historical landslide disaster area data to obtain the landslide sensitivity assessment model. The assessment module uses the landslide sensitivity assessment model to assess the landslide sensitivity of the target area and obtains the landslide sensitivity assessment results for the target area. The partitioning module is specifically used to perform the following steps: Initialize the sparrow population for: in, This represents the value of the catastrophic factor, a continuous variable, in the first dimension for the first sparrow. Indicates the first The sparrow only considers the values ​​of the catastrophic factors, which are continuous variables, in the first dimension. This indicates that the first sparrow was in the [number]th [location]. The values ​​of the catastrophic factors of the continuous variables mentioned in the document. Indicates the first Only sparrows in the first The values ​​of the disaster-causing factors of the continuous variables mentioned in the document; Update # The location of the sparrow is determined by the formula: Update the discoverer's location; in, Indicates the first Only sparrows in the first Wei Zhongyu Position at the next iteration Indicates the first Only sparrows in the first Wei Zhongyu Position at the next iteration , This represents the total number of sparrows in the sparrow population. , This represents the total number of dimensions. Indicates the maximum number of iterations. It is a random number. , Indicates the warning value. Indicates a safe value. Represents a random number. Represents the parameter matrix; Through the formula: Update the location of the newcomer; in, Indicates the first The worst position in the dimension. Indicates the discoverer was in the During the nth iteration, at the... The best position in the dimensional, Represents the parameter matrix; Through the formula: Update the guards' positions; in, Indicates the first The globally optimal position in the dimension. Indicates the step control parameters. Indicates the current fitness. This represents the globally optimal fitness. It is a random number. , Indicates the worst global fitness. It is a constant; The first Only sparrows in the first Wei Zhongyu Position at the next iteration As the continuous variable causative factor The Group secondary factors The second-order factor and the first The interval values ​​between the secondary factors; Through the formula: Calculate the continuous variable hazard factor In the corresponding continuous variable hazard factor layer, the first Normalized values ​​of landslide density in the regions corresponding to each secondary factor ; in, The continuous variable hazard factor The The area of ​​the region corresponding to each second-level factor The continuous variable hazard factor The The historical landslide area in the region corresponding to each secondary factor , This represents the total number of disaster-causing factors among the continuous variables; Through the objective function: Calculate the continuous variable hazard factor The Entropy of group second-order factors ; The continuous variable catastrophic factors The Entropy of group second-order factors As the first The adaptability of a sparrow Get the fitness of all sparrows. : in, This indicates the fitness level of the first sparrow. Indicates the first The adaptability of a single sparrow; Determine whether the number of iterations has reached the maximum number of iterations. If the maximum number of iterations has not been reached, then... The fitness with the smallest value is taken as the global optimal fitness, and the position of the sparrow corresponding to the smallest fitness value is taken as the global best position. The fitness with the largest median fitness value is taken as the global worst fitness value. The position of the sparrow corresponding to the fitness with the largest median fitness value is taken as the global worst position. The global optimal fitness value, the global worst fitness value, the global best position, and the global worst position are updated, and the update is returned. The steps to locate the sparrow; If the maximum number of iterations is reached, then the iteration stops, and the value at which the maximum number of iterations was reached is recorded. As the continuous variable causative factor The The second-order factor and the first The interval values ​​between the secondary factors; Based on the secondary factors of the continuous variable disaster-causing factors, the continuous variable disaster-causing factor layer is divided to obtain the secondary factor division layer of the continuous variable disaster-causing factors.

7. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the landslide sensitivity assessment method based on second-level factor interval optimization as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the landslide sensitivity assessment method based on the second-level factor interval optimization division as described in any one of claims 1 to 5.

Images