Rainfall landslide probability risk assessment method, device, equipment and program product

Abstract

This invention discloses a method, apparatus, equipment, and program product for risk assessment of rainfall-induced landslide probability. It constructs a probability density evolution model based on historical rainfall data of the target area; determines the critical threshold for triggering cluster landslides under different rainfall durations based on the probability density evolution model and the acceptable risk probability threshold; constructs a risk assessment model based on historical cluster landslide cataloging data of the target area, the critical threshold for triggering cluster landslides, and disaster-prone element data; constructs a random rainfall scenario, and uses the risk assessment model to assess the probability of cluster landslides under the random rainfall scenario. This invention solves the problem of poor adaptability of traditional static thresholds by determining rainfall thresholds through a probability density evolution model; it achieves quantitative characterization of the spatiotemporal uncertainty of the geological environment by integrating disaster-prone elements and rainfall thresholds to construct a risk assessment model; and it improves the accuracy of early warning by dynamically assessing the probability of future cluster landslides in conjunction with random rainfall scenarios.

Description

Technical Field

[0001] This invention relates to the field of risk prediction technology, specifically to risk assessment methods, devices, equipment, and procedures for assessing the probability of landslides caused by rainfall. Background Technology

[0002] Landslides are a typical form of geological hazard, with rain-induced landslides being particularly notable for their suddenness, wide impact, and destructive power. Current rain-induced landslide risk assessment techniques primarily employ single probability distribution fitting or empirical statistical models to describe the correlation between rainfall conditions, geological environment, and landslide occurrence. These methods typically analyze historical rainfall data and landslide logging data to establish threshold curves for rainfall intensity versus duration, or utilize statistical regression methods to assess the landslide susceptibility of specific areas, providing a foundation for geological hazard early warning and prevention.

[0003] However, the relevant technologies have significant limitations. First, due to the complexity and spatial heterogeneity of the geological environment, as well as the randomness and uncertainty of rainfall events, traditional single probability distribution fitting methods are insufficient to accurately characterize the critical conditions for rainfall-triggered landslides. This results in rainfall thresholds that are often too static and empirical, failing to adapt to the dynamic changes of different rainfall processes. Second, traditional methods lack effective means to quantitatively characterize the spatiotemporal uncertainty probabilities of disaster-causing factors, making it difficult to dynamically integrate rainfall thresholds with multiple factors such as topography, geology, and hydrology. This leads to significant discrepancies between risk assessment results and actual cluster landslide disasters. Summary of the Invention

[0004] This invention provides a risk assessment method, apparatus, equipment, and program product for rainfall-induced landslide probability, which solves the problem of how to accurately determine the critical threshold for rainfall-triggered landslides and realize the quantitative characterization of the spatiotemporal uncertainty probability of disaster-causing factors, thereby breaking through the limitations of traditional static threshold methods and realizing the dynamic assessment of the risk of rainfall-induced cluster landslides.

[0005] In a first aspect, the present invention provides a risk assessment method for the probability of rainfall-induced landslides, the method comprising: Obtain historical rainfall data for the target area, and construct a probability density evolution model based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations; Based on the probability density evolution model and the preset risk acceptable probability threshold, the critical threshold for triggering cluster landslides in the target area under different rainfall durations is determined. Acquire historical mass landslide catalog data and disaster-prone element data for the target area. Based on the historical mass landslide catalog data, the rainfall-triggered mass landslide critical threshold, and the disaster-prone element data, construct a risk assessment model to characterize the probability of mass landslides occurring under the combined effect of rainfall threshold and disaster-prone elements. A random rainfall scenario for the target area is constructed, and the risk assessment model is used to assess the probability of cluster landslides under the random rainfall scenario.

[0006] This invention determines rainfall thresholds through a probability density evolution model, solving the problem of poor adaptability of traditional static thresholds; it integrates disaster-prone factors with rainfall thresholds to construct a risk assessment model, realizing a quantitative characterization of the spatiotemporal uncertainty of the geological environment; and it combines random rainfall scenarios to dynamically assess the probability of future cluster landslides, improving the accuracy of early warning.

[0007] In one optional implementation, the step of constructing a probability density evolution model based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations includes: Using the rainfall duration in the historical rainfall data as the evolution parameter and the cumulative rainfall as the state variable, a partial differential equation is established to express the probability density of the cumulative rainfall as a function of the rainfall duration. The probability evolution rate in the partial differential equation is determined by regression analysis on the historical rainfall data. Solving the partial differential equation yields the probability density function of the cumulative rainfall under different rainfall durations, and the probability density evolution model is constructed based on the probability density function.

[0008] This invention solves the problem that traditional static threshold methods struggle to describe the dynamic evolution of rainfall probability distribution over time by constructing partial differential equations using rainfall duration as the evolution parameter and cumulative rainfall as the state variable. By performing regression analysis on historical rainfall data to determine the probability evolution rate, it achieves a true reconstruction of the statistical patterns of rainfall in the target area. Furthermore, by solving the partial differential equations, it obtains the probability density function of cumulative rainfall under different rainfall durations, constructing a complete probability density evolution model. This enables an accurate description of the dynamic characteristics of rainfall processes, effectively improving the rationality and reliability of risk assessments for rainfall-induced cluster landslides.

[0009] In one alternative implementation, the probability evolution rate is expressed by the following formula:

[0010] In the formula, , , These are the fitted parameters estimated using the least squares method; This refers to the duration of rainfall. This refers to the cumulative rainfall over the duration of the rainfall. The probability evolution rate of cumulative rainfall is used to reflect the trend of the probability distribution of cumulative rainfall as the duration of rainfall increases.

[0011] In one optional implementation, determining the critical threshold for triggering cluster landslides in the target area under different rainfall durations, based on the probability density evolution model and a preset acceptable risk probability threshold, includes: Integrating the probability density function corresponding to the cumulative rainfall, we obtain the probability of triggering a cluster landslide when the cumulative rainfall is reached or exceeded during the rainfall duration. The minimum cumulative rainfall amount that makes the probability of triggering the mass landslide equal to a preset risk acceptable probability threshold is determined, and the minimum cumulative rainfall amount is determined as the critical threshold for triggering the mass landslide corresponding to the rainfall duration.

[0012] This invention solves the problem of traditional static threshold methods failing to convert statistical patterns of rainfall into quantifiable risk indicators by integrating the probability density function corresponding to cumulative rainfall to obtain the probability of triggering landslides in clusters. It determines the minimum cumulative rainfall that makes the probability of triggering a cluster landslide equal to a preset acceptable risk threshold as the critical threshold for triggering a cluster landslide. Furthermore, it allows for flexible setting of early warning triggering conditions based on the risk tolerance of the target area, transforming the mathematical output of the probability density evolution model into an early warning indicator, thus providing reliable basic data input for subsequent risk assessment of cluster landslides.

[0013] In one optional implementation, the step of constructing a risk assessment model based on the historical landslide cataloging data, the rainfall-triggered landslide threshold, and the disaster-prone factor data to characterize the probability of landslide occurrence under the combined effect of rainfall threshold and disaster-prone factors includes: Compare the cumulative rainfall in the historical rainfall data with the critical threshold for triggering a cluster of landslides under the corresponding rainfall duration; If the cumulative rainfall reaches or exceeds the critical threshold for triggering a landslide, the rainfall condition variable is assigned a first state; if the cumulative rainfall does not reach the critical threshold for triggering a landslide, the rainfall condition variable is assigned a second state. Using the assigned rainfall conditional variable and the disaster-prone element as independent variables, and the historical cluster landslide catalog data as dependent variable, a risk assessment model is constructed to characterize the probability of cluster landslides occurring under the combined effect of rainfall threshold and disaster-prone element.

[0014] This invention solves the problem of traditional methods failing to integrate dynamic rainfall thresholds and static disaster-prone factors into a single risk assessment framework by comparing historical cumulative rainfall with rainfall-triggered landslide thresholds and converting the comparison into binary rainfall conditional variables. By constructing a logistic regression model using the assigned rainfall conditional variables and disaster-prone factors as independent variables and historical landslide cataloging data as the dependent variable, the invention achieves an organic fusion of rainfall thresholds and disaster-prone factors. This allows for a quantitative characterization of the contribution of rainfall conditions reaching the threshold to the probability of landslide occurrence, taking into account both the dynamic characteristics of rainfall and the inherent properties of the geological environment when assessing the risk of landslide clusters.

[0015] In one optional implementation, constructing the random rainfall scenario for the target area includes: Historical rainfall statistical features of the target area are extracted based on the historical rainfall data. Based on the aforementioned historical rainfall statistics, a random rainfall scenario is constructed that includes rainfall duration and cumulative rainfall.

[0016] This invention extracts historical rainfall statistical features of a target area based on historical rainfall data, and constructs random rainfall scenarios that include rainfall duration and cumulative rainfall based on these statistical features. This solves the problem that traditional deterministic rainfall scenario settings cannot cover various possible future rainfall patterns. By generating a large number of random rainfall samples that conform to historical rainfall patterns through random sampling methods, it achieves a comprehensive quantitative characterization of the uncertainty of future rainfall conditions in the target area. It can generate diverse rainfall scenarios covering short-duration light rain to long-duration heavy rain, providing a comprehensive input data foundation for subsequent risk assessment of the probability of landslides.

[0017] In one optional implementation, the step of using the risk assessment model to assess the probability of cluster landslides under the random rainfall scenario includes: For each of the aforementioned random rainfall scenarios, the target rainfall duration and target cumulative rainfall amount for that random rainfall scenario are obtained; Based on the target rainfall duration, determine the target rainfall triggering mass landslide critical threshold corresponding to the random rainfall scenario, and determine whether the target cumulative rainfall reaches or exceeds the target rainfall triggering mass landslide critical threshold, so as to determine the target rainfall condition variable corresponding to the random rainfall scenario; The target rainfall condition variable and the disaster-prone element data are input into the risk assessment model to obtain the probability of target cluster landslides occurring under the random rainfall scenario. Statistical analysis was performed on the probability of landslide occurrence in the target area corresponding to all random rainfall scenarios to obtain the probability distribution of landslide risk in the target area.

[0018] This invention addresses the problem of traditional risk assessment methods failing to effectively combine dynamic rainfall thresholds with random rainfall scenarios by obtaining target rainfall duration and target cumulative rainfall for each random rainfall scenario. It determines a corresponding critical threshold based on the target rainfall duration and assesses whether the cumulative rainfall reaches the threshold to identify rainfall condition variables. By inputting rainfall condition variables and disaster-prone element data into the risk assessment model to calculate the probability of landslide occurrence, and statistically analyzing the calculation results for all scenarios to obtain the landslide risk probability distribution, this invention achieves a quantitative assessment of the risk of clustered landslides under uncertain future rainfall conditions. It comprehensively reflects the landslide risk level faced by the target area under different rainfall scenarios.

[0019] Secondly, the present invention provides a risk assessment device for the probability of landslides caused by rainfall, the device comprising: The evolution model construction and threshold determination module is used to acquire historical rainfall data of the target area, construct a probability density evolution model based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations, and determine the critical threshold for triggering cluster landslides in the target area under different rainfall durations according to the probability density evolution model and the preset risk-acceptable probability threshold. The assessment model construction module is used to acquire historical cluster landslide catalog data and disaster-prone element data of the target area. Based on the historical cluster landslide catalog data, the rainfall-triggered cluster landslide critical threshold, and the disaster-prone element data, a risk assessment model is constructed to characterize the probability of cluster landslides occurring under the combined effect of rainfall threshold and disaster-prone elements. The risk assessment module is used to construct a random rainfall scenario for the target area and use the risk assessment model to assess the probability of cluster landslides under the random rainfall scenario.

[0020] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the risk assessment method for rainfall-induced landslide probability described in the first aspect or any corresponding embodiment thereof.

[0021] Fourthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the risk assessment method for rainfall-induced landslide probability described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0022] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the first process of a risk assessment method for the probability of landslides caused by rainfall according to an embodiment of the present invention; Figure 3 This is a structural block diagram of a risk assessment device for the probability of landslides caused by rainfall according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0026] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0027] As an optional application scenario of this invention, such as Figure 1 As shown, the risk assessment system for the probability of landslides due to rainfall may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0028] Specifically, the terminal device can be a smartphone, tablet, laptop, or PDA. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranets, local area networks, wide area networks, mobile communication networks, and combinations thereof. It should be noted that... Figure 1 This is merely an example of an application scenario and does not limit the scope of protection of this invention.

[0029] Landslides are a typical form of geological hazard, with rain-induced landslides being particularly notable for their suddenness, wide impact, and destructive power. Current rain-induced landslide risk assessment techniques primarily employ single probability distribution fitting or empirical statistical models to describe the correlation between rainfall conditions, geological environment, and landslide occurrence. These methods typically analyze historical rainfall data and landslide logging data to establish threshold curves for rainfall intensity versus duration, or utilize statistical regression methods to assess the landslide susceptibility of specific areas, providing a foundation for geological hazard early warning and prevention.

[0030] However, the relevant technologies have significant limitations. First, due to the complexity and spatial heterogeneity of the geological environment, as well as the randomness and uncertainty of rainfall events, traditional single probability distribution fitting methods are insufficient to accurately characterize the critical conditions for rainfall-triggered landslides. This results in determined rainfall thresholds that are often too static and empirical, failing to adapt to the dynamic changes of different rainfall processes. Second, traditional methods lack effective means to quantitatively characterize the spatiotemporal uncertainty probabilities of disaster-causing factors, making it difficult to dynamically integrate rainfall factors with multiple factors such as topography, geology, and hydrology. This leads to significant discrepancies between risk assessment results and actual cluster landslide disasters.

[0031] Based on this, the present invention provides an embodiment of a risk assessment method for the probability of landslides caused by rainfall. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0032] This embodiment provides a risk assessment method for the probability of landslides caused by rainfall, which can be used in the risk assessment system described above. Figure 2 This is a flowchart of a risk assessment method for rainfall-induced landslide probability according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: Obtain historical rainfall data for the target area, and construct a probability density evolution model based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations.

[0033] Step S202: Based on the probability density evolution model and the preset acceptable risk probability threshold, determine the critical threshold for triggering cluster landslides in the target area under different rainfall durations.

[0034] Step S203: Obtain historical cluster landslide catalog data and disaster-prone element data for the target area. Based on the historical cluster landslide catalog data, the critical threshold for triggering cluster landslides by rainfall, and the disaster-prone element data, construct a risk assessment model to characterize the probability of cluster landslides occurring under the combined effect of rainfall threshold and disaster-prone elements.

[0035] Step S204: Construct a random rainfall scenario for the target area and use a risk assessment model to assess the probability of cluster landslides under the random rainfall scenario.

[0036] The risk assessment method for rainfall-induced landslide probability provided in this embodiment constructs a probability density evolution model based on historical rainfall data of the target area; determines the critical threshold for triggering cluster landslides under different rainfall durations based on the probability density evolution model and the acceptable risk probability threshold; constructs a risk assessment model based on historical cluster landslide cataloging data of the target area, the critical threshold for triggering cluster landslides, and disaster-prone element data; constructs a random rainfall scenario, and uses the risk assessment model to assess the probability of cluster landslides under the random rainfall scenario. This embodiment solves the problem of poor adaptability of traditional static thresholds by determining the rainfall threshold through the probability density evolution model; it achieves quantitative characterization of the spatiotemporal uncertainty of the geological environment by integrating disaster-prone elements and rainfall thresholds to construct the risk assessment model; and it improves the accuracy of early warning by dynamically assessing the probability of future cluster landslides in conjunction with random rainfall scenarios.

[0037] The steps described above will be explained in detail below.

[0038] In step S201, historical rainfall data of the target area is obtained, and a probability density evolution model is constructed based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations.

[0039] In one embodiment, a probability density evolution model is constructed based on historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations, including: Using rainfall duration from historical rainfall data as the evolution parameter and cumulative rainfall as the state variable, a partial differential equation is established to show the probability density of cumulative rainfall as a function of rainfall duration. The probability evolution rate in the partial differential equation is determined by regression analysis on historical rainfall data. Solving the partial differential equation yields the probability density function of cumulative rainfall under different rainfall durations, and a probability density evolution model is constructed based on the probability density function.

[0040] This embodiment uses rainfall duration from historical rainfall data as the basic parameter controlling the evolution, and cumulative rainfall as the core variable describing the system state. Based on this, a partial differential equation is constructed that expresses how the probability density of cumulative rainfall continuously changes with increasing rainfall duration. In the partial differential equation, the specific numerical value or functional form of the probability evolution rate is obtained through regression analysis of historical rainfall data. Then, the established partial differential equation is mathematically solved to calculate the probability density function corresponding to the cumulative rainfall under different rainfall duration conditions. Finally, using this series of probability density functions, the probability density evolution model is constructed.

[0041] In this context, evolution parameters refer to variables that control the change of probability distribution over time or process; in this embodiment, it refers to rainfall duration. State variables are variables that describe the state of the system at a certain moment; in this embodiment, it refers to cumulative rainfall. Probability density is a measure of the likelihood of a random variable occurring near a certain value point. Partial differential equations are equations containing unknown functions and their partial derivatives, used to describe the changing patterns of multivariable functions. Probability evolution rate is a key component of partial differential equations, used to express how quickly the probability distribution of cumulative rainfall changes with increasing rainfall duration. Regression analysis is a statistical analysis method that determines the relationship between variables using historical data. The probability density function is a function that describes the probability density of a random variable at each possible value point.

[0042] Specifically, firstly, based on the long-term risk zoning of regional geological hazards, high-risk areas are identified as target areas, and historical rainfall data for these areas is obtained. Historical rainfall data originates from long-term accumulated observation records from meteorological observation stations or rainfall monitoring networks in the target area. This data includes daily rainfall, rainfall duration, and process records of individual rainfall events. Specifically, daily rainfall data can be statistically recorded on a calendar day basis, with the total rainfall amount per day measured in millimeters. Rainfall duration refers to the length of time a rainfall event lasts from start to finish, measured in days. For rainfall events with a duration exceeding one day, the daily rainfall amounts from multiple consecutive days need to be accumulated to form a complete rainfall event record. Simultaneously, the scope of historical rainfall data collection should cover a sufficiently long observation period to ensure that the data reflects the statistical regularity of rainfall in the target area. After collection, historical rainfall data needs to undergo data cleaning and outlier removal, such as excluding abnormal rainfall data caused by instrument malfunction or recording errors, thereby ensuring the reliability of the data used for subsequent modeling.

[0043] After acquiring and preprocessing historical rainfall data, a probability density evolution model is constructed based on this data. First, each historical rainfall event needs to be categorized according to its duration to form a cumulative rainfall sample set for different durations. For each rainfall duration category, such as one day, two days, three days, up to the longest observed rainfall duration, a corresponding cumulative rainfall sample sequence is compiled. Cumulative rainfall refers to the sum of all daily rainfall within that duration, expressed in millimeters. For example, for a two-day rainfall event, the cumulative rainfall equals the sum of the rainfall on the first and second days. Based on the obtained cumulative rainfall sample sets for different durations, a partial differential equation is established, using rainfall duration as the evolution parameter and cumulative rainfall as the state variable, to express the probability density of cumulative rainfall as a function of rainfall duration.

[0044] When constructing the probability density evolution model, the probability density evolution theory is first used to construct a multi-timescale critical rainfall threshold probability characterization model based on cumulative rainfall and daily rainfall data. The probability density evolution equation is a formula for fitting the distribution of probability density with duration, used to find the probabilistic pattern of cumulative rainfall occurrence under different rainfall durations. The specific expression of the probability density evolution equation is as follows:

[0045] In the formula, The duration of rainfall is calculated in days. This is the sum of all daily rainfall during the duration of the rainfall, i.e., the cumulative rainfall, expressed in millimeters. The probability evolution rate of cumulative rainfall is used to reflect the trend of the probability distribution of cumulative rainfall as the duration of rainfall increases; Let be the probability density function, which is the rate of change of the probability of a landslide triggering when the cumulative rainfall increases under a given rainfall duration. At that time, the cumulative rainfall was exactly The probability of it.

[0046] The probability evolution rate is expressed by the following formula:

[0047] In the formula, , , These are the fitted parameters estimated using the least squares method; This refers to the duration of rainfall. This refers to the cumulative rainfall over the duration of the rainfall. The probability evolution rate of cumulative rainfall is used to reflect the trend of the probability distribution of cumulative rainfall as the duration of rainfall increases.

[0048] This embodiment solves the problem that traditional static threshold methods struggle to describe the dynamic evolution of rainfall probability distribution over time by constructing partial differential equations using rainfall duration as the evolution parameter and cumulative rainfall as the state variable. By performing regression analysis on historical rainfall data to determine the probability evolution rate, it achieves a true reconstruction of the statistical patterns of rainfall in the target area. Furthermore, by solving the partial differential equations, it obtains the probability density function of cumulative rainfall under different rainfall durations, constructing a complete probability density evolution model. This enables an accurate description of the dynamic characteristics of rainfall processes, effectively improving the rationality and reliability of risk assessments for rainfall-induced cluster landslides.

[0049] In step S202, based on the probability density evolution model and the preset acceptable risk probability threshold, the critical threshold for triggering mass landslides in the target area under different rainfall durations is determined.

[0050] In one embodiment, based on a probability density evolution model and a preset acceptable risk probability threshold, the critical threshold for triggering cluster landslides in the target area under different rainfall durations is determined, including: Integrating the probability density function corresponding to the cumulative rainfall, we obtain the probability of triggering a cluster of landslides when the cumulative rainfall is reached or exceeded during the rainfall duration. Determine the minimum cumulative rainfall amount that makes the probability of triggering a cluster of landslides equal to the preset acceptable risk probability threshold, and set the minimum cumulative rainfall amount as the critical threshold for triggering a cluster of landslides corresponding to the rainfall duration.

[0051] This embodiment first integrates the probability density function corresponding to the cumulative rainfall to obtain the probability of a landslide triggering when the cumulative rainfall reaches or exceeds a certain value for a given rainfall duration. The probability density function describes the likelihood of different cumulative rainfall amounts occurring, while landslide risk prevention requires the actual probability of a landslide occurring when the cumulative rainfall reaches a certain level. Therefore, by integrating the probability density function from a certain cumulative rainfall value to infinity, the probability of a landslide triggering when the cumulative rainfall reaches or exceeds that value under that rainfall duration condition can be obtained. Different values ​​of the lower bound of integration result in different landslide trigger probabilities; the higher the cumulative rainfall threshold, the lower the corresponding landslide trigger probability, and vice versa.

[0052] Next, the minimum cumulative rainfall amount required to make the probability of triggering a cluster of landslides equal to a preset acceptable risk probability threshold is determined, and this minimum cumulative rainfall amount is defined as the critical threshold for triggering a cluster of landslides for that rainfall duration. The preset acceptable risk probability threshold is used to characterize the tolerance of the target area to landslide risk. For example, when the acceptable risk probability threshold is set to 80%, it means that it is expected that an early warning or preventive measures will be initiated when the probability of triggering a cluster of landslides reaches 80%. Through iterative calculation, the minimum cumulative rainfall amount that satisfies the integral result equal to the preset threshold can be determined, and thus this cumulative rainfall amount is the critical threshold for triggering a cluster of landslides for that rainfall duration. By performing the above calculations for different rainfall durations, a complete set of rainfall threshold curves covering various scenarios from short-duration rainfall to long-duration rainfall can be obtained.

[0053] Among them, the landslide triggering probability refers to the probability that a landslide will actually occur when the cumulative rainfall reaches or exceeds a certain value under a given rainfall duration; the acceptable risk probability threshold refers to the landslide triggering probability limit value pre-set according to the actual disaster prevention needs and risk tolerance of the target area, which is used as the judgment benchmark for back-calculating the critical threshold of rainfall-triggered landslides; the critical threshold of rainfall-triggered cluster landslides refers to the minimum cumulative rainfall that makes the landslide triggering probability exactly equal to the acceptable risk probability threshold under a specific rainfall duration, which is the core indicator used for early warning of rainfall-induced cluster landslides.

[0054] Specifically, when constructing a probability density evolution model, initial and boundary conditions need to be set. Since the integral of the probability density function equals 1 when covering the entire value space (meaning the sum of the probability density integrals for all possible cumulative rainfall amounts is 1), and the only possible value for cumulative rainfall in the absence of rainfall is zero, the probability density is concentrated at the zero point, taking a value of one. The physical meaning of this initial condition is that the probability density is one when there is no rainfall, but the actual probability of triggering a landslide is zero. This is because zero cumulative rainfall in the absence of rainfall obviously will not trigger a landslide event. Therefore, the initial conditions can be set as zero rainfall duration, zero cumulative rainfall, and a probability density value of one.

[0055] The boundary condition is set such that when the cumulative rainfall reaches or exceeds a preset extreme value, the probability density value approaches zero, thereby ensuring that the probability density function has reasonable convergence behavior when the cumulative rainfall is extremely large.

[0056] After setting the initial and boundary conditions, an integral transformation is performed on the probability density function output by the probability density evolution model. For a given rainfall duration, the probability density function is integrated over an interval from a certain cumulative rainfall value to infinity. The integral result represents the probability of triggering a cluster of landslides when the cumulative rainfall reaches or exceeds that value under that rainfall duration condition. The specific integration formula is as follows:

[0057] In the formula, This refers to the duration of rainfall. This refers to the cumulative rainfall over the duration of the rainfall. This is the cumulative rainfall threshold used to determine whether a landslide has been triggered given a given rainfall duration. Let be the probability density function, which is the rate of change of the probability of a landslide triggering when the cumulative rainfall increases under a given rainfall duration. At that time, the cumulative rainfall was exactly The probability of; To the duration of rainfall Below, the cumulative rainfall reached or exceeded The probability of triggering a landslide.

[0058] For example , hour, This indicates that if there is continuous rainfall for two days and the cumulative rainfall reaches or exceeds 100mm, there is an 80% probability of triggering a landslide risk.

[0059] Conversely, by setting a preset acceptable risk probability threshold, such as requiring an early warning and preventative measures to be taken when the preset acceptable risk probability threshold reaches or exceeds 80%, the cumulative rainfall corresponding to each rainfall duration can be calculated:

[0060] In the formula, This refers to the duration of rainfall. This refers to the cumulative rainfall over the duration of the rainfall. This is the cumulative rainfall threshold used to determine whether a landslide has been triggered given a given rainfall duration. Let be the probability density function, which is the rate of change of the probability of a landslide triggering when the cumulative rainfall increases under a given rainfall duration. At that time, the cumulative rainfall was exactly The probability of; This represents the acceptable probability threshold for risk.

[0061] This embodiment solves the problem that traditional static threshold methods struggle to transform rainfall statistics into quantifiable risk indicators by integrating the probability density function corresponding to cumulative rainfall. It establishes a critical threshold for triggering landslides based on rainfall, ensuring the probability of triggering a landslide equals a preset acceptable risk threshold. Furthermore, it allows for flexible setting of early warning triggering conditions based on the risk tolerance of the target area. This transforms the mathematical output of the probability density evolution model into early warning indicators, providing reliable basic data input for subsequent landslide risk assessment.

[0062] In step S203, historical landslide cataloging data and disaster-prone element data of the target area are obtained. Based on the historical landslide cataloging data, the critical threshold for triggering landslides by rainfall, and the disaster-prone element data, a risk assessment model is constructed to characterize the probability of landslides occurring under the combined effect of rainfall threshold and disaster-prone elements.

[0063] In one embodiment, a risk assessment model is constructed based on historical landslide cataloging data, rainfall-triggered landslide thresholds, and disaster-prone factor data to characterize the probability of landslides occurring under the combined effects of rainfall thresholds and disaster-prone factors. The model includes: The cumulative rainfall in historical rainfall data is compared with the critical threshold for triggering a landslide at the corresponding rainfall duration. If the cumulative rainfall reaches or exceeds the critical threshold for triggering a landslide, the rainfall condition variable is assigned the first state. If the cumulative rainfall does not reach the critical threshold for triggering a landslide, the rainfall condition variable is assigned the second state. Using the assigned rainfall conditional variable and disaster-prone factors as independent variables, and historical mass landslide cataloging data as dependent variables, a risk assessment model is constructed to characterize the probability of mass landslides occurring under the combined effect of rainfall threshold and disaster-prone factors.

[0064] In this embodiment, historical landslide cataloging data and disaster-prone element data for the target area are first acquired. The historical landslide cataloging data records information on landslide events that have occurred in the target area throughout history, including the time, location, and scale of the landslides. Disaster-prone element data refers to the inherent environmental attributes that determine the susceptibility to geological disasters in the target area. It reflects the inherent attributes of the geological environment of the target area and can include multiple aspects such as topographic factors, geological condition factors, hydrological environmental factors, and land cover factors. Specifically, disaster-prone elements can include indicators such as elevation, slope, aspect, surface relief, topographic humidity index, lithology, vegetation cover, soil type, land cover type, and distance from rivers. Specifically, the disaster-prone element data is processed using a geographic information system to extract attributes from the disaster-prone element data to landslide locations, ensuring that each landslide point is matched with a corresponding disaster-prone element attribute value.

[0065] After obtaining historical landslide cataloging data and disaster-prone element data, the cumulative rainfall in the historical rainfall data is compared with the rainfall-triggered landslide threshold for the corresponding rainfall duration. For each historical rainfall event, the rainfall duration of the historical rainfall event is first determined, then the rainfall-triggered landslide threshold corresponding to that rainfall duration is determined, and the actual cumulative rainfall of the current historical rainfall event is compared with the rainfall-triggered landslide threshold. If the actual cumulative rainfall reaches or exceeds the rainfall-triggered landslide threshold, the rainfall condition variable is assigned a first state, for example, a value of 1, to indicate that the rainfall conditions have reached the trigger threshold; if the actual cumulative rainfall does not reach the rainfall-triggered landslide threshold, the rainfall condition variable is assigned a second state, for example, a value of 0, to indicate that the rainfall conditions have not reached the trigger threshold.

[0066] A risk assessment model was constructed using the assigned rainfall conditional variable and disaster-prone element attribute values ​​as independent variables and historical mass landslide catalog data as dependent variables. The dependent variable was in binary form, with events that actually occurred being assigned a value of 1 and events that did not occur being assigned a value of 0.

[0067] Next, a logistic regression algorithm is used to establish the mapping relationship between independent and dependent variables. The logistic regression model calculates the weighted linear combination value of each influencing factor and obtains the landslide probability through a logistic function transformation. This model can be expressed as a linear combination formula, where each influencing factor includes the attribute values ​​of disaster-prone elements and the regression coefficients corresponding to rainfall condition variables. These coefficients are solved using the maximum likelihood estimation method based on the training set data. The magnitude of the regression coefficients reflects the contribution of each influencing factor to the landslide probability. A larger regression coefficient for the rainfall condition variable indicates a stronger influence of rainfall reaching the threshold condition on triggering a landslide. The mapping relationship is as follows:

[0068] In the formula, The probability of a landslide occurring is between [0, 1]. This is the weighted linear composite value of all influencing factors; For each influencing factor; These are the logistic regression coefficients; This is a rainfall threshold condition factor; These are the logistic regression coefficients corresponding to the rainfall threshold condition factor. The larger the value, the stronger the impact of rainfall on landslides. It can be used to quantitatively evaluate the sensitivity of landslides in a target area to rainfall conditions.

[0069] In the process of building a risk assessment model, historical landslide cataloging data can be divided into training and validation sets according to a predetermined ratio. For example, an 8:2 ratio can be used, with 80% of the samples used for model training and 20% for model validation. The training set refers to the data sample set selected from historical data to fit the model parameters, which can determine the regression coefficients of each influencing factor. The validation set refers to the data sample set reserved from historical data that was not used in model training, which can be used to test the model's fit to known data. An initial risk assessment model is constructed by fitting logistic regression coefficients using the training set data. The completed risk assessment model can characterize the quantitative relationship between the probability of landslides occurring under the combined effects of rainfall threshold conditions and disaster-causing factors, thereby dynamically assessing the probability risk of landslides in future scenarios.

[0070] This embodiment solves the problem of traditional methods struggling to integrate dynamic rainfall thresholds and static disaster-prone factors into a single risk assessment framework by comparing historical cumulative rainfall with rainfall-triggered landslide thresholds and converting the comparisons into binary rainfall conditional variables. By constructing a logistic regression model using the assigned rainfall conditional variables and disaster-prone factors as independent variables and historical landslide cataloging data as the dependent variable, the embodiment achieves the organic integration of rainfall thresholds and disaster-prone factors. This allows for a quantitative characterization of the contribution of rainfall conditions reaching the threshold to the probability of landslide occurrence, taking into account both the dynamic characteristics of rainfall and the inherent properties of the geological environment when assessing the risk of landslide clusters.

[0071] Furthermore, the accuracy of the constructed probability prediction model for clustered landslides can be validated, including two levels of validation. Since the samples in the validation set were not involved in the model training process, the first level of validation uses a reserved validation set from the historical clustered landslide catalog data for internal validation. During validation, the disaster-prone factor data and rainfall condition variables of each sample in the validation set are input into the trained risk assessment model to calculate the predicted landslide probability value for each sample. The model prediction results (i.e., the predicted landslide probability values ​​in this embodiment) are compared sample-by-sample with the actual landslide occurrences recorded in the validation set. Based on the comparison results, a receiver operating characteristic (ROC) curve is plotted. The ROC curve is a curve plotted with the false positive rate on the x-axis and the true positive rate on the y-axis, used to comprehensively reflect the model's classification performance under different classification thresholds. The area under the ROC curve is calculated, with the value ranging from [0.5, 1]. The closer the area is to 1, the more reliable the model's classification performance; an area of ​​0.5 indicates that the model's prediction effect is equivalent to random guessing. When the calculated area value reaches the preset reliability standard, it indicates that the model has good accuracy in internal validation.

[0072] The second level of validation involves using a new historical rainfall event and its corresponding cluster of landslides for external validation. This historical rainfall event and landslide data were not used in the model building and internal validation processes and are completely independent of the modeling process. The purpose of external validation is to test the model's generalization ability, i.e., its predictive performance when applied to new data. During validation, the disaster-causing factors and rainfall condition variables from the external validation data are input into the constructed risk assessment model to obtain landslide probability predictions. The prediction results (i.e., the landslide probability predictions in this embodiment) are then compared with the actual cluster of landslides. Based on the comparison results, a receiver operating characteristic (ROC) curve is plotted, and the area under the ROC curve is calculated as the external validation accuracy. The external validation accuracy is compared with the first level of validation accuracy. If the external validation accuracy is close to the internal validation accuracy and both meet the preset reliability standards, it indicates that the model has good generalization ability. If the external validation accuracy is significantly lower than the internal validation accuracy, it indicates that the model is overfitting and needs optimization. Overfitting refers to the phenomenon where a model performs well on training data but poorly on new data, usually due to an overly complex model or insufficient training data.

[0073] When model optimization is required, parameter adjustments and structural optimizations can be performed on the risk assessment model based on the validation results to further improve its accuracy. Parameter adjustments include re-estimating regression coefficients or optimizing the selection and combination of disaster risk factors. Structural optimization includes adjusting the ratio of the training and validation sets or introducing regularization methods to prevent overfitting. Through multiple iterations of validation and optimization, the model achieves stable accuracy metrics in both internal and external validation, ultimately determining the optimal parameter configuration for the model.

[0074] In step S204, a random rainfall scenario for the target area is constructed, and a risk assessment model is used to assess the probability of cluster landslides under the random rainfall scenario.

[0075] In one embodiment, constructing a random rainfall scenario for the target area includes: Historical rainfall statistics are extracted for the target area based on historical rainfall data; based on these historical rainfall statistics, a random rainfall scenario including rainfall duration and cumulative rainfall is constructed.

[0076] Historical rainfall data can be derived from long-term accumulated observation records from meteorological observation stations in the target area, including daily rainfall, rainfall duration, and complete records of each rainfall event. Statistical analysis of historical rainfall data extracts historical rainfall statistical features that reflect the rainfall patterns in the target area. These features can include the probability distribution of rainfall duration, the probability distribution of cumulative rainfall, and the joint distribution relationship between rainfall duration and cumulative rainfall. Specifically, the frequency distribution of different rainfall durations in historical records is determined, and extreme value analysis is used to determine the characteristic values ​​of cumulative rainfall at different return periods. Correlation analysis is then used to determine the correlation coefficient between rainfall duration and cumulative rainfall.

[0077] Based on the extracted historical rainfall statistical features, a random rainfall scenario including rainfall duration and cumulative rainfall is constructed. The construction process employs random sampling to generate a large number of random rainfall samples that conform to the rainfall patterns of the target area, based on the historical rainfall statistical features.

[0078] For each random rainfall sample, a rainfall duration value is first randomly selected based on the probability distribution of historical rainfall durations. This rainfall duration value represents the duration of the simulated rainfall event, measured in days. After determining the rainfall duration, a cumulative rainfall value is randomly selected based on the conditional probability distribution of the cumulative rainfall under that rainfall duration. This cumulative rainfall value represents the simulated total rainfall within that rainfall duration. The random selection processes for rainfall duration and cumulative rainfall are interconnected, ensuring that the generated random rainfall scenarios accurately reflect the inherent coupling relationship between rainfall duration and intensity in the target area. By repeating the above random sampling process, a sufficient number of random rainfall scenarios are generated, covering various possible rainfall patterns from short-duration light rain to long-duration torrential rain, comprehensively reflecting the future rainfall conditions that the target area may face.

[0079] This embodiment extracts historical rainfall statistical features of the target area based on historical rainfall data, and constructs random rainfall scenarios that include rainfall duration and cumulative rainfall based on these statistical features. This solves the problem that traditional deterministic rainfall scenario settings cannot cover various possible future rainfall patterns. By generating a large number of random rainfall samples that conform to historical rainfall patterns through random sampling methods, it achieves a comprehensive quantitative characterization of the uncertainty of future rainfall conditions in the target area. It can generate diverse rainfall scenarios covering short-duration light rain to long-duration heavy rain, providing a comprehensive input data foundation for subsequent risk assessment of the probability of mass landslides.

[0080] In one embodiment, a risk assessment model is used to assess the probability of cluster landslides under a random rainfall scenario, including: For each random rainfall scenario, obtain the target rainfall duration and target cumulative rainfall for that random rainfall scenario; Based on the target rainfall duration, determine the critical threshold for triggering a cluster of landslides corresponding to the random rainfall scenario, and determine whether the target cumulative rainfall reaches or exceeds the critical threshold for triggering a cluster of landslides, so as to determine the target rainfall condition variables corresponding to the random rainfall scenario; By inputting the target rainfall condition variables and disaster-prone factor data into the risk assessment model, the probability of target cluster landslides occurring under random rainfall scenarios is obtained. Statistical analysis was performed on the probability of landslides occurring in the target area under all random rainfall scenarios to obtain the probability distribution of landslide risk in the target area.

[0081] This embodiment uses the generated random rainfall scenarios as input to the risk assessment model to dynamically assess the probability of cluster landslides under future scenarios. Each set of random rainfall scenarios includes a defined rainfall duration and cumulative rainfall amount, which can be compared with the corresponding rainfall-triggered cluster landslide critical threshold to determine whether the rainfall condition variable in this scenario should be assigned a first state or a second state, and then input into the risk assessment model to calculate the probability of the target cluster landslide occurring under this scenario.

[0082] First, for each generated random rainfall scenario, the target rainfall duration and target cumulative rainfall are obtained. Random rainfall scenarios are generated using a random sampling method. Each random rainfall scenario includes a defined rainfall duration value (i.e., the target rainfall duration in this embodiment) and a corresponding cumulative rainfall value (i.e., the target cumulative rainfall in this embodiment). The target rainfall duration is in days and represents the duration of the simulated rainfall event; the target cumulative rainfall is in millimeters and represents the simulated total rainfall within that rainfall duration.

[0083] After obtaining the target rainfall duration and target cumulative rainfall, the critical threshold for triggering a cluster of landslides corresponding to the random rainfall scenario is determined based on the target rainfall duration. Specifically, the probability density function corresponding to the target cumulative rainfall is integrated to obtain the probability of triggering a cluster of landslides when the target cumulative rainfall is reached or exceeded under the target rainfall duration; the minimum target cumulative rainfall that makes the probability of triggering a cluster of landslides equal to the preset acceptable risk probability threshold is determined, and the minimum target cumulative rainfall is determined as the critical threshold for triggering a cluster of landslides corresponding to the target rainfall duration.

[0084] The target cumulative rainfall is compared with the target rainfall-triggered landslide threshold. If the target cumulative rainfall reaches or exceeds the target rainfall-triggered landslide threshold, the target rainfall condition variable is assigned a first state, such as a value of one. If the target cumulative rainfall does not reach the target rainfall-triggered landslide threshold, the target rainfall condition variable is assigned a second state, such as a value of zero, so that rainfall factors can participate in subsequent risk assessments in the form of discrete variables.

[0085] After determining the target rainfall condition variable, the target rainfall condition variable and the disaster-prone element data of the target area are input into the risk assessment model. The risk assessment model is constructed using a logistic regression algorithm. After inputting the target rainfall condition variable and the disaster-prone element data, the probability of target cluster landslides under this random rainfall scenario is obtained. The probability of target cluster landslides ranges from [0, 1], reflecting the possibility of cluster landslides occurring under the combined effects of rainfall conditions and geological environment in this set of random rainfall scenarios.

[0086] The above processing procedure is repeated for all random rainfall scenarios to obtain the probability of landslide occurrence for each random rainfall scenario. After calculating the probability of landslide occurrence for all random rainfall scenarios, statistical analysis is performed on the probability of landslide occurrence for all target groups. The statistical analysis includes calculating statistical characteristics such as the mean, standard deviation, and quantiles of the landslide occurrence probability, plotting a frequency distribution histogram or cumulative probability distribution curve of the landslide occurrence probability, and identifying the occurrence frequency of different probability intervals. Through statistical analysis, the probability distribution of landslide risk in the target area under future rainfall conditions is obtained. This probability distribution comprehensively reflects the landslide risk level faced by the target area under different rainfall scenarios.

[0087] This embodiment obtains the target rainfall duration and target cumulative rainfall for each random rainfall scenario, determines the corresponding critical threshold based on the target rainfall duration, and judges whether the cumulative rainfall reaches the threshold to determine the rainfall condition variable. This solves the problem that traditional risk assessment methods are difficult to effectively combine dynamic rainfall thresholds with random rainfall scenarios. By inputting the rainfall condition variable and disaster-prone element data into the risk assessment model to calculate the probability of landslide occurrence, and performing statistical analysis on the calculation results of all scenarios to obtain the landslide risk probability distribution, a quantitative assessment of the risk of cluster landslides under uncertain future rainfall conditions is achieved. This can comprehensively reflect the landslide risk level faced by the target area under different rainfall scenarios.

[0088] This embodiment also provides a risk assessment device for the probability of landslides caused by rainfall. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0089] This embodiment provides a risk assessment device for the probability of landslides caused by rainfall, such as... Figure 3 As shown, it includes: The evolution model construction and threshold determination module 301 is used to acquire historical rainfall data of the target area, construct a probability density evolution model based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations, and determine the critical threshold for triggering cluster landslides in the target area under different rainfall durations according to the probability density evolution model and the preset risk acceptable probability threshold.

[0090] The assessment model construction module 302 is used to acquire historical cluster landslide catalog data and disaster-prone element data of the target area. Based on the historical cluster landslide catalog data, the critical threshold for triggering cluster landslides by rainfall, and the disaster-prone element data, a risk assessment model is constructed to characterize the probability of cluster landslides occurring under the combined effect of rainfall threshold and disaster-prone elements.

[0091] The risk assessment module 303 is used to construct random rainfall scenarios for the target area and to use a risk assessment model to assess the probability of cluster landslides under random rainfall scenarios.

[0092] In some optional implementations, the evolution model construction and threshold determination module 301 is specifically used to establish a partial differential equation for the probability density of cumulative rainfall as a function of rainfall duration, using rainfall duration in historical rainfall data as the evolution parameter and cumulative rainfall as the state variable. The probability evolution rate in the partial differential equation is determined by regression analysis of historical rainfall data. The partial differential equation is solved to obtain the probability density function of cumulative rainfall under different rainfall durations, and a probability density evolution model is constructed based on the probability density function.

[0093] In some alternative implementations, the probability evolution rate is expressed by the following formula:

[0094] In the formula, , , These are the fitted parameters estimated using the least squares method; This refers to the duration of rainfall. This refers to the cumulative rainfall over the duration of the rainfall. The probability evolution rate of cumulative rainfall is used to reflect the trend of the probability distribution of cumulative rainfall as the duration of rainfall increases.

[0095] In some optional implementations, the evolution model construction and threshold determination module 301 is specifically used to integrate the probability density function corresponding to the cumulative rainfall to obtain the probability of triggering a group landslide when the cumulative rainfall is reached or exceeded under the rainfall duration; determine the minimum cumulative rainfall that makes the probability of triggering a group landslide equal to the preset risk acceptable probability threshold, and determine the minimum cumulative rainfall as the rainfall triggering group landslide critical threshold corresponding to the rainfall duration.

[0096] In some optional implementations, the assessment model construction module 302 is specifically used to compare the cumulative rainfall in historical rainfall data with the rainfall-triggered cluster landslide critical threshold for the corresponding rainfall duration; if the cumulative rainfall reaches or exceeds the rainfall-triggered cluster landslide critical threshold, the rainfall condition variable is assigned a first state; if the cumulative rainfall does not reach the rainfall-triggered cluster landslide critical threshold, the rainfall condition variable is assigned a second state; using the assigned rainfall condition variable and disaster-prone factors as independent variables, and historical cluster landslide cataloging data as dependent variables, a risk assessment model is constructed to characterize the probability of cluster landslides occurring under the combined effect of rainfall threshold and disaster-prone factors.

[0097] In some optional implementations, the risk assessment module 303 is specifically used to extract historical rainfall statistical features of the target area based on historical rainfall data; and to construct a random rainfall scenario that includes rainfall duration and cumulative rainfall based on the historical rainfall statistical features.

[0098] In some optional implementations, the risk assessment module 303 is specifically used to obtain the target rainfall duration and target cumulative rainfall for each random rainfall scenario; determine the critical threshold for triggering a cluster landslide based on the target rainfall duration, and determine whether the target cumulative rainfall reaches or exceeds the critical threshold for triggering a cluster landslide, so as to determine the target rainfall condition variable corresponding to the random rainfall scenario; input the target rainfall condition variable and disaster-prone element data into the risk assessment model to obtain the probability of the target cluster landslide under the random rainfall scenario; and perform statistical analysis on the probability of the target cluster landslide corresponding to all random rainfall scenarios to obtain the probability distribution of landslide risk in the target area.

[0099] The risk assessment device for rainfall-induced landslide probability provided in this embodiment of the invention can execute the risk assessment method for rainfall-induced landslide probability provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0100] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0101] The following is a detailed reference. Figure 4 This diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 401, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 402 or a program loaded from memory 408 into random access memory (RAM) 403. The RAM 403 also stores various programs and data required for the operation of the electronic device. The processor 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0102] Typically, the following devices can be connected to I / O interface 405: input devices 406 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 408 including, for example, magnetic tapes, hard disks, etc.; and communication devices 409. Communication device 409 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0103] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 409, or installed from a memory 408, or installed from a ROM 402. When the computer program is executed by the processor 401, it performs the functions defined in the risk assessment method for rainfall-induced landslide probability according to embodiments of the present invention.

[0104] Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0105] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the risk assessment method for rainfall-induced landslide probability shown in the above embodiments is implemented.

[0106] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0107] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A risk assessment method for the probability of rainfall-induced landslides, characterized in that, The method includes: Obtain historical rainfall data for the target area, and construct a probability density evolution model based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations; Based on the probability density evolution model and the preset risk acceptable probability threshold, the critical threshold for triggering cluster landslides in the target area under different rainfall durations is determined. Acquire historical mass landslide catalog data and disaster-prone element data for the target area. Based on the historical mass landslide catalog data, the rainfall-triggered mass landslide critical threshold, and the disaster-prone element data, construct a risk assessment model to characterize the probability of mass landslides occurring under the combined effect of rainfall threshold and disaster-prone elements. A random rainfall scenario for the target area is constructed, and the risk assessment model is used to assess the probability of cluster landslides under the random rainfall scenario.

2. The method according to claim 1, characterized in that, The probability density evolution model constructed based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations includes: Using the rainfall duration in the historical rainfall data as the evolution parameter and the cumulative rainfall as the state variable, a partial differential equation is established to express the probability density of the cumulative rainfall as a function of the rainfall duration. The probability evolution rate in the partial differential equation is determined by regression analysis on the historical rainfall data. Solving the partial differential equation yields the probability density function of the cumulative rainfall under different rainfall durations, and the probability density evolution model is constructed based on the probability density function.

3. The method according to claim 2, characterized in that, The probability evolution rate is expressed by the following formula: In the formula, , , These are the fitted parameters estimated using the least squares method; This refers to the duration of rainfall. This refers to the cumulative rainfall over the duration of the rainfall. The probability evolution rate of cumulative rainfall is used to reflect the trend of the probability distribution of cumulative rainfall as the duration of rainfall increases.

4. The method according to claim 2, characterized in that, The step of determining the critical threshold for triggering cluster landslides in the target area under different rainfall durations based on the probability density evolution model and a preset acceptable risk probability threshold includes: Integrating the probability density function corresponding to the cumulative rainfall, we obtain the probability of triggering a cluster landslide when the cumulative rainfall is reached or exceeded during the rainfall duration. The minimum cumulative rainfall amount that makes the probability of triggering the mass landslide equal to a preset risk acceptable probability threshold is determined, and the minimum cumulative rainfall amount is determined as the critical threshold for triggering the mass landslide corresponding to the rainfall duration.

5. The method according to claim 1, characterized in that, The risk assessment model, constructed based on the historical landslide cataloging data, the rainfall-triggered landslide threshold, and the disaster-prone factor data, is used to characterize the probability of landslide occurrence under the combined effect of rainfall threshold and disaster-prone factors. The model includes: Compare the cumulative rainfall in the historical rainfall data with the critical threshold for triggering a cluster of landslides under the corresponding rainfall duration; If the cumulative rainfall reaches or exceeds the critical threshold for triggering a landslide, the rainfall condition variable is assigned a first state; if the cumulative rainfall does not reach the critical threshold for triggering a landslide, the rainfall condition variable is assigned a second state. Using the assigned rainfall conditional variable and the disaster-prone element as independent variables, and the historical cluster landslide catalog data as dependent variable, a risk assessment model is constructed to characterize the probability of cluster landslides occurring under the combined effect of rainfall threshold and disaster-prone element.

6. The method according to claim 1, characterized in that, The construction of the random rainfall scenario for the target area includes: Historical rainfall statistical features of the target area are extracted based on the historical rainfall data. Based on the aforementioned historical rainfall statistics, a random rainfall scenario is constructed that includes rainfall duration and cumulative rainfall.

7. The method according to claim 6, characterized in that, The process of using the risk assessment model to assess the probability of cluster landslides under the random rainfall scenario includes: For each of the aforementioned random rainfall scenarios, the target rainfall duration and target cumulative rainfall amount for that random rainfall scenario are obtained; Based on the target rainfall duration, determine the target rainfall triggering mass landslide critical threshold corresponding to the random rainfall scenario, and determine whether the target cumulative rainfall reaches or exceeds the target rainfall triggering mass landslide critical threshold, so as to determine the target rainfall condition variable corresponding to the random rainfall scenario; The target rainfall condition variable and the disaster-prone element data are input into the risk assessment model to obtain the probability of target cluster landslides occurring under the random rainfall scenario. Statistical analysis was performed on the probability of landslide occurrence in the target area corresponding to all random rainfall scenarios to obtain the probability distribution of landslide risk in the target area.

8. A risk assessment device for the probability of landslides caused by rainfall, characterized in that, The device includes: The evolution model construction and threshold determination module is used to acquire historical rainfall data of the target area, construct a probability density evolution model based on the historical rainfall data to describe the probability distribution of cumulative rainfall for different rainfall durations, and determine the critical threshold for triggering cluster landslides in the target area under different rainfall durations according to the probability density evolution model and the preset risk-acceptable probability threshold. The assessment model construction module is used to acquire historical cluster landslide catalog data and disaster-prone element data of the target area. Based on the historical cluster landslide catalog data, the rainfall-triggered cluster landslide critical threshold, and the disaster-prone element data, a risk assessment model is constructed to characterize the probability of cluster landslides occurring under the combined effect of rainfall threshold and disaster-prone elements. The risk assessment module is used to construct a random rainfall scenario for the target area and use the risk assessment model to assess the probability of cluster landslides under the random rainfall scenario.

9. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the risk assessment method for rainfall-induced landslide probability as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the risk assessment method for rainfall-induced landslide probability as described in any one of claims 1 to 7.

Images