A Critical Early Warning Method and System for Debris Flow Initiation Based on Dynamic Weights of Multiple Influencing Parameters

Abstract

This invention provides a method and system for early warning of debris flow initiation based on dynamic weighting of multiple influencing parameters, belonging to the field of geological disaster prevention and early warning technology. The method includes: constructing a set of multiple influencing parameters by acquiring source conditions, topographic features, and real-time rainfall parameters; constructing an intrinsic weighting function based on the initiation physical mechanism to calculate the intrinsic weights of the source and topographic parameters; adjusting the intrinsic weights using a dynamic correction function with rainfall intensity or cumulative rainfall as input to obtain dynamic weights; establishing a critical function coupling the dynamic weights and critical conditions to calculate the debris flow initiation critical index, and issuing an early warning when the threshold is exceeded. This solves the problem that existing debris flow early warning methods are unable to simultaneously integrate physical mechanisms and dynamic weight adjustment, and accurately reflect the critical state of debris flow initiation under the coupling effect of multiple factors.

Description

Technical Field

[0001] This invention relates to the field of geological disaster prevention and early warning technology, and more specifically, to a method and system for early warning of debris flow initiation based on dynamic weights of multiple influencing parameters. Background Technology

[0002] Debris flow early warning is one of the core challenges in the field of geological disaster prevention and control. Existing technologies mainly focus on rainfall as a key triggering factor. For example, patent document CN106652361A proposes a debris flow early warning method based on rainfall-probability. This method establishes a rainfall-probability early warning index system by determining the rainfall threshold for debris flow occurrence and combining it with conditional probability models under different rainfall patterns. This method reflects the uncertainty of debris flow occurrence to a certain extent and improves the scientific nature of the early warning. However, this technology still uses rainfall as the core indicator and does not adequately consider other key factors affecting debris flow initiation, such as source conditions and topographic features. Its probability model mainly relies on the statistical relationship between historical rainfall and disaster events, making it difficult to dynamically reflect the complex mechanism of multi-factor coupling in different debris flow basin environments.

[0003] On the other hand, with the development of artificial intelligence technology, debris flow forecasting methods based on multi-factor data-driven approaches have gradually attracted attention. Patent document CN111967648A proposes a debris flow disaster forecasting method based on a width learning model. This method collects multiple influencing factors such as rainfall, slope gradient, and soil moisture content through a sensor network, and uses a fast principal component extraction algorithm for dimensionality reduction. Finally, it outputs the probability of debris flow occurrence through a width learning model. This method overcomes the limitations of a single rainfall indicator and can handle multi-source heterogeneous data. However, this technology is essentially a data-driven black box model. Its prediction results depend on the quality and coverage of the training data, lacking a clear characterization of the physical initiation mechanism of debris flows, especially the inability to dynamically adjust the weights of each influencing factor under critical conditions. This limits its applicability and interpretability under different geological environments or rainfall conditions.

[0004] In summary, existing debris flow early warning methods either focus on rainfall indicators while neglecting multi-factor coupling mechanisms, or rely on data-driven models but lack the ability to understand physical mechanisms and dynamically adjust weights. These methods struggle to accurately reflect the critical state of debris flow initiation under the combined effects of different material source conditions, topographic features, and rainfall processes. Therefore, there is an urgent need to develop a debris flow initiation critical early warning method that can integrate physical mechanisms and multi-source information, and achieve dynamic weight adjustment of influencing factors. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for early warning of debris flow initiation based on dynamic weights of multiple influencing parameters. This aims to solve the problem that existing debris flow early warning methods are unable to simultaneously integrate physical mechanisms and dynamic weight adjustments, and accurately reflect the critical state of debris flow initiation under the coupled effects of multiple factors.

[0006] This invention is achieved through the following technical solution: A critical early warning method for debris flow initiation based on dynamic weights of multiple influencing parameters includes the following steps: Obtain source condition parameters, topographic feature parameters, and real-time rainfall process parameters of the target watershed, and construct a set of multiple influence parameters; Based on the physical mechanism of debris flow initiation, an intrinsic weight function reflecting the contribution of source conditions and topographic features to debris flow initiation is constructed, and the intrinsic weights of each influencing parameter are calculated according to the source condition parameters and topographic feature parameters. Based on real-time rainfall process parameters, the intrinsic weights are adjusted using a dynamic correction function to obtain the dynamic weights of each influencing parameter; wherein, the dynamic correction function takes rainfall intensity or cumulative rainfall as input and weight adjustment coefficient as output, and is used to reflect the time-varying enhancement or weakening effect of the rainfall process on the sensitivity of the source conditions and topographic features; Establish a critical function that couples dynamic weights with the critical condition for debris flow initiation, substitute the weighted set of multiple influence parameters into the critical function, and calculate the critical index for debris flow initiation under the current working condition. When the debris flow initiation critical index exceeds a preset threshold, a debris flow initiation warning message is issued.

[0007] Optionally, the specific process of obtaining the source condition parameters, topographic feature parameters, and real-time rainfall process parameters of the target watershed, and constructing a multi-influence parameter set, is as follows: The source conditions parameters of the target watershed are obtained through on-site investigation and remote sensing interpretation. These source conditions parameters include the reserves of solid materials that can participate in the initiation of debris flows, the material gradation characteristics, and the internal friction angle of the materials. Topographic feature parameters are obtained through digital elevation modeling and field surveying. These parameters include the longitudinal gradient of the gully bed, the catchment area of ​​the watershed, and the morphological ratio of the gully width to its depth. Rainfall process parameters are collected in real time through a rainfall monitoring network deployed in the target watershed. These parameters include rainfall intensity, effective cumulative rainfall, and rainfall duration over a given period. The parameters of the material source conditions, the parameters of the terrain features, and the parameters of the real-time rainfall process are parameterized according to a unified time scale and spatial resolution, forming a multi-influence parameter set that includes the feature vectors of the material source conditions, the feature vectors of the terrain, and the feature vectors of the rainfall process.

[0008] Optionally, the specific process of constructing an intrinsic weighting function reflecting the contribution of source conditions and topographic features to debris flow initiation based on the physical mechanism of debris flow initiation, and calculating the intrinsic weights of each influencing parameter based on source condition parameters and topographic feature parameters, is as follows: The solid material reserves, material gradation characteristics, and material internal friction angle in the material source condition parameters are parameterized as material source supply factor, gradation characteristic factor, and shear strength factor, respectively; the gully bed longitudinal gradient, watershed catchment area, and gully morphology ratio in the topographic characteristic parameters are parameterized as topographic driving factor, runoff capacity factor, and gully constraint factor, respectively. Based on the physical coupling mechanism of sediment supply and topographic driving during debris flow initiation, an intrinsic weight function is constructed with sediment supply factor, gradation characteristic factor, shear strength factor, topographic driving factor, runoff capacity factor, and channel constraint factor as independent variables. The intrinsic weight function uses a combination of the analytic hierarchy process (AHP) and the entropy weight method to determine the inherent contribution of each factor in debris flow initiation. Specifically, the AHP constructs a judgment matrix based on the physical mechanism of debris flow initiation, which is used to determine the subjective weight of each factor. The entropy weight method calculates the objective weight of each factor based on historical sediment supply conditions and topographic feature data of the target watershed. The intrinsic weight of each factor is obtained by multiplying and normalizing the subjective and objective weights. The intrinsic weight vector, which characterizes the inherent contribution of each influencing parameter, is obtained by multiplying the material supply factor, gradation characteristic factor, shear strength factor, terrain driving factor, runoff capacity factor, and channel constraint factor by their respective intrinsic weights.

[0009] Optionally, the specific process of adjusting the intrinsic weights using a dynamic correction function based on real-time rainfall process parameters to obtain the dynamic weights of each influencing parameter is as follows: The dynamic correction function is defined as a two-parameter piecewise function with time-period rainfall intensity and effective cumulative rainfall as inputs. The dynamic correction function includes a source sensitivity correction sub-function and a topography sensitivity correction sub-function. The source sensitivity correction sub-function is used to characterize the time-varying enhancement or weakening effect of the rainfall process on the intrinsic weights corresponding to the source condition parameters, and the topography sensitivity correction sub-function is used to characterize the time-varying enhancement or weakening effect of the rainfall process on the intrinsic weights corresponding to the topography feature parameters. The real-time collected rainfall intensity and effective cumulative rainfall are simultaneously input into the source sensitivity correction sub-function and the terrain sensitivity correction sub-function, and the weight adjustment coefficients corresponding to each source condition parameter and each terrain feature parameter are calculated respectively. The intrinsic weights of each influencing parameter in the material source condition parameters and the intrinsic weights of each influencing parameter in the terrain feature parameters are multiplied by their corresponding weight adjustment coefficients to obtain the dynamic weights of each material source condition parameter and the dynamic weights of each terrain feature parameter. The dynamic weight vector of multiple influencing parameters is composed of the dynamic weights of the various source condition parameters and the dynamic weights of the various topographic feature parameters.

[0010] Optionally, both the source sensitivity correction sub-function and the terrain sensitivity correction sub-function adopt a piecewise nonlinear form. When the effective cumulative rainfall is lower than or equal to the lower limit of the first threshold interval, the weight adjustment coefficient increases linearly with the increase of rainfall intensity over time. When the effective cumulative rainfall is higher than the upper limit of the first threshold interval and lower than or equal to the lower limit of the second threshold interval, the weight adjustment coefficient increases exponentially with the increase of rainfall intensity over time. When the effective cumulative rainfall exceeds the upper limit of the second threshold interval, the weight adjustment coefficient reaches the saturation upper limit and no longer changes with the increase of rainfall intensity over time; The values ​​of the first threshold interval and the second threshold interval are determined based on the rainfall process data of historical debris flow events in the target watershed, as well as the sensitivity analysis results of the corresponding source conditions and topographic features, and are simultaneously reviewed and corrected when the intrinsic weights of the target watershed are updated.

[0011] Optionally, the specific process of establishing the critical function that couples the dynamic weights with the critical condition for debris flow initiation is as follows: Based on the physical mechanism of debris flow initiation, the basic form of the critical function is a ratio function of the weighted driving term and the critical resistance term. The weighted driving term is used to characterize the comprehensive driving capacity of the material source conditions and terrain features on debris flow initiation under dynamic weighting. The critical resistance term is used to characterize the critical resistance conditions that need to be overcome for debris flow initiation under the current working conditions. The source condition parameters and terrain feature parameters in the multi-influence parameter set are multiplied by their corresponding dynamic weights to obtain a weighted source parameter vector and a weighted terrain parameter vector, respectively. The weighted source parameter vector and the weighted terrain parameter vector are then summed element by element to obtain a comprehensive driving index. Based on the source conditions and topographic features of the target watershed, a critical resistance term is constructed using the limit equilibrium analysis method; wherein, the critical resistance term includes the resistance component provided by the shear strength of the source and the resistance component provided by the channel geometric constraints; The ratio of the comprehensive driving index to the critical resistance term is defined as the debris flow initiation critical exponent, and the critical function is established as shown in the following formula:

[0012] in, The critical index for the initiation of debris flows; For the first The weight values ​​of each influencing parameter under dynamic weighting; For the first Normalized values ​​of the influencing parameters; The number of influencing parameters; For the first One resistance component; This represents the quantity of the resistance component.

[0013] Optionally, the specific process of issuing a debris flow initiation warning when the debris flow initiation critical index exceeds a preset threshold is as follows: Based on the limit equilibrium critical conditions for debris flow initiation in the target watershed, measured samples of critical indices corresponding to historical disaster events, and statistical samples of critical indices for historical conditions without disasters, a multi-level progressive early warning threshold sequence corresponding to the risk level of debris flow initiation is calibrated. The multi-level progressive early warning threshold sequence is divided into three threshold levels based on the critical equilibrium state of the driving and resistance terms corresponding to the critical function: low-risk early warning threshold, medium-risk early warning threshold, and high-risk early warning threshold. The values ​​of each threshold level are synchronously reviewed and corrected as the intrinsic weights of the impact parameters of the target watershed are updated. According to the time step consistent with the real-time rainfall process parameters, the debris flow initiation critical index is updated and calculated synchronously. The debris flow initiation critical index calculated in real time is compared with the multi-level progressive early warning threshold sequence level by level to determine the debris flow risk level corresponding to the current working condition. When the real-time calculated debris flow initiation critical index exceeds the warning threshold of the corresponding level, a threshold exceedance persistence check is initiated. The threshold exceedance persistence check involves continuously checking the debris flow initiation critical index for no less than three preset time steps. If the debris flow initiation critical index is continuously higher than the warning threshold of the corresponding level throughout the entire check period, the warning triggering condition of the corresponding level is determined to be met. If the debris flow initiation critical index falls below the corresponding warning threshold during the check period, the check is terminated and the warning of the corresponding level is not triggered. Once the warning triggering conditions are met, based on the dynamic weight vector of each influencing parameter under the current working conditions, the contribution ratio of each influencing parameter to the debris flow initiation critical index is calculated, the top two dominant influencing parameters with the highest contribution ratio are identified, and warning information including debris flow risk level, dominant influencing parameters, real-time rainfall conditions, and the time series change trend of the critical index is generated. The warning information is released according to the preset warning release level and release rules. The calculation and update time step of the debris flow initiation critical index is compressed to half of the original step step, and encrypted calculation and dynamic tracking are performed until the debris flow initiation critical index falls back below the low-risk warning threshold.

[0014] Based on the same inventive concept, this invention also provides a debris flow initiation critical early warning system based on dynamic weights of multiple influencing parameters, used to implement the aforementioned debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters, comprising: The parameter acquisition module is configured to acquire source condition parameters, terrain feature parameters, and real-time rainfall process parameters through field surveys, remote sensing interpretation, digital elevation models, field mapping, and rainfall monitoring networks deployed in the target watershed. The acquired parameters are then parameterized according to a unified time scale and spatial resolution to construct a multi-influence parameter set containing source condition feature vectors, terrain feature vectors, and rainfall process feature vectors. The intrinsic weight calculation module, whose input end is connected to the output end of the parameter acquisition module, is configured as follows: based on the physical mechanism of debris flow initiation, an intrinsic weight function reflecting the contribution of source conditions and terrain features to debris flow initiation is constructed; according to the source condition parameters and terrain feature parameters output by the parameter acquisition module, the intrinsic weight of each influencing parameter is calculated by combining the analytic hierarchy process and the entropy weight method, and the intrinsic weight vector is output. A dynamic weight correction module, whose input is connected to the output of the parameter acquisition module and the output of the intrinsic weight calculation module, is configured to: calculate the weight adjustment coefficients corresponding to each source condition parameter and each terrain feature parameter respectively using a dynamic correction function that includes a source sensitivity correction sub-function and a terrain sensitivity correction sub-function, based on the real-time rainfall process parameters output by the parameter acquisition module; dynamically adjust the intrinsic weight vector output by the intrinsic weight calculation module using the weight adjustment coefficients; and output a dynamic weight vector of multiple influencing parameters. The critical index calculation module, whose input is connected to the output of the parameter acquisition module and the output of the dynamic weight correction module, is configured to: construct a critical function that couples dynamic weights with the critical conditions for debris flow initiation based on the physical mechanism of debris flow initiation; substitute the set of multiple influence parameters output by the parameter acquisition module and the dynamic weight vector output by the dynamic weight correction module into the critical function; calculate the critical index for debris flow initiation under the current working condition; and output the critical index for debris flow initiation. The early warning release module, whose input is connected to the output of the critical index calculation module, is configured to: compare the debris flow initiation critical index output by the critical index calculation module with a preset multi-level progressive early warning threshold sequence; and generate and release early warning information including debris flow risk level, dominant influencing parameters, real-time rainfall conditions, and the time-series change trend of the critical index when the debris flow initiation critical index is continuously higher than the corresponding level's early warning threshold for at least three consecutive preset time steps.

[0015] Based on the same inventive concept, the present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the above-described debris flow initiation critical early warning method based on dynamic weights of multiple influence parameters.

[0016] Based on the same inventive concept, the present invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described debris flow initiation critical early warning method based on dynamic weights of multiple influence parameters.

[0017] The technical solution of the present invention has at least the following advantages and beneficial effects: By constructing an intrinsic weight function that reflects the contribution of material source conditions and terrain features to the initiation of debris flows, the physical mechanism of debris flow initiation is integrated into the early warning model, giving the weight configuration of influencing factors a clear physical meaning and enhancing the interpretability and scientific validity of the early warning results.

[0018] By introducing a dynamic correction function, the weights of each influencing factor are dynamically adjusted according to real-time rainfall process parameters (such as rainfall intensity and cumulative rainfall). This can reflect the time-varying enhancement or weakening effect of the rainfall process on the sensitivity of source conditions and topographic features, enabling the model to reasonably adjust the contribution of each factor under different rainfall conditions, thereby improving the adaptability and accuracy of the early warning method in complex rainfall environments.

[0019] By comprehensively considering three key parameters—material source conditions, topographic features, and rainfall processes—a critical function coupled with dynamic weights and debris flow initiation critical conditions is constructed. This function can more comprehensively reflect the driving mechanism of debris flow initiation caused by the combined effects of multiple factors under different watershed environments. Compared with single index or static weight models, it improves the accuracy of identifying the critical state of debris flow initiation. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a debris flow initiation critical early warning system based on dynamic weights of multiple influencing parameters, according to an embodiment of the present invention. Detailed Implementation

[0021] The following is a detailed description of the embodiments, in conjunction with the accompanying drawings.

[0022] Reference Figure 1 A critical early warning method for debris flow initiation based on dynamic weights of multiple influencing parameters includes the following steps: Step 1: Obtain the source conditions, topographic features, and real-time rainfall parameters of the target watershed, and construct a set of multiple influencing parameters.

[0023] In some embodiments, the specific process of obtaining the source condition parameters, topographic feature parameters, and real-time rainfall process parameters of the target watershed, and constructing a multi-influence parameter set is as follows: The source conditions parameters of the target watershed are obtained through on-site investigation and remote sensing interpretation. These source conditions parameters include the reserves of solid materials that can participate in the initiation of debris flows, the material gradation characteristics, and the internal friction angle of the materials. Topographic feature parameters are obtained through digital elevation modeling and field surveying. These parameters include the longitudinal gradient of the gully bed, the catchment area of ​​the watershed, and the morphological ratio of the gully width to its depth. Rainfall process parameters are collected in real time through a rainfall monitoring network deployed in the target watershed. These parameters include rainfall intensity, effective cumulative rainfall, and rainfall duration over a given period. The parameters of the material source conditions, the parameters of the terrain features, and the parameters of the real-time rainfall process are parameterized according to a unified time scale and spatial resolution, forming a multi-influence parameter set that includes the feature vectors of the material source conditions, the feature vectors of the terrain, and the feature vectors of the rainfall process.

[0024] Step 2: Based on the physical mechanism of debris flow initiation, construct an intrinsic weight function that reflects the contribution of source conditions and topographic features to debris flow initiation, and calculate the intrinsic weight of each influencing parameter according to the source condition parameters and topographic feature parameters.

[0025] In some embodiments, the specific process of constructing an intrinsic weighting function reflecting the contribution of source conditions and topographic features to debris flow initiation based on the physical mechanism of debris flow initiation, and calculating the intrinsic weights of each influencing parameter based on source condition parameters and topographic feature parameters, is as follows: The solid material reserves, material gradation characteristics, and material internal friction angle in the material source condition parameters are parameterized as material source supply factor, gradation characteristic factor, and shear strength factor, respectively; the gully bed longitudinal gradient, watershed catchment area, and gully morphology ratio in the topographic characteristic parameters are parameterized as topographic driving factor, runoff capacity factor, and gully constraint factor, respectively. Based on the physical coupling mechanism of sediment supply and topographic driving during debris flow initiation, an intrinsic weight function is constructed with sediment supply factor, gradation characteristic factor, shear strength factor, topographic driving factor, runoff capacity factor, and channel constraint factor as independent variables. The intrinsic weight function uses a combination of the analytic hierarchy process (AHP) and the entropy weight method to determine the inherent contribution of each factor in debris flow initiation. Specifically, the AHP constructs a judgment matrix based on the physical mechanism of debris flow initiation, which is used to determine the subjective weight of each factor. The entropy weight method calculates the objective weight of each factor based on historical sediment supply conditions and topographic feature data of the target watershed. The intrinsic weight of each factor is obtained by multiplying and normalizing the subjective and objective weights. The intrinsic weight vector, which characterizes the inherent contribution of each influencing parameter, is obtained by multiplying the material supply factor, gradation characteristic factor, shear strength factor, terrain driving factor, runoff capacity factor, and channel constraint factor by their respective intrinsic weights.

[0026] We can define the set of key influencing factors for debris flow initiation as follows: ,in: As a factor of material supply; It is the gradation characteristic factor; It is the shear strength factor; Terrain-driven factors; This is the confluence capacity factor; This is the channel constraint factor.

[0027] Judgment matrix constructed based on the physical mechanism of debris flow The definition is as follows:

[0028] in, ; To determine the matrix The Middle Line number The elements of the column represent factors. Relative factor The importance scale for the contribution to debris flow initiation is assigned a value based on the causal strength in the physical mechanism.

[0029] To ensure that the judgment matrix conforms to the physical logic of debris flow initiation, The value of follows these rules:

[0030] in, and They represent the first Factors and the Factors The contribution quantification in the physical initiation mechanism of debris flows is determined by a physical function based on the limit equilibrium theory, as shown in the following formula:

[0031] in, The critical index for debris flow initiation under reference working conditions; As a factor Sensitivity to the critical index under reference operating conditions; The physical constraint factor is used to reflect the causal hierarchy of the factor in the initiation mechanism (e.g., the material supply factor is a direct supply mechanism, and the terrain driving factor is an energy driving mechanism).

[0032] Substituting the above importance scale based on physical mechanisms into the following judgment matrix :

[0033] According to the analytic hierarchy process, from the judgment matrix The subjective weight vector is calculated as follows:

[0034] in, To determine the matrix The subjective weight vector is obtained by normalizing the largest eigenvalue and the corresponding eigenvector. ,in: Subjective weighting of material supply factors; The subjective weights of the gradation characteristic factors; The subjective weight of the shear strength factor; Subjective weights for terrain-driving factors; The subjective weight of the convergence capacity factor; The subjective weight of the channel constraint factor.

[0035] The target watershed can be set to have a total of A set of historical sediment source conditions and topographic features data samples, each sample containing 6 factors (sediment supply factors). Gradation characteristic factor Shear strength factor Terrain-driven factors Convergence capacity factor Channel constraint factor Construct the original data matrix. ,in Indicates the first In the nth sample The values ​​of each factor, The number of impact factors, .

[0036] The range normalization method is used to normalize the original data matrix. The normalization process is performed as shown in the following formula:

[0037] in, Indicates the first In the nth sample The normalized values ​​of each factor, the normalized matrix is .

[0038] Based on the normalized matrix Calculate the first The factor under the th The proportion of each sample is shown in the following formula:

[0039] in, Indicates the first The factor under the th The proportion of each sample; if Then let To avoid undefined logarithms.

[0040] Calculate the percentage based on the proportion. The information entropy of each factor is shown in the following formula:

[0041] in, Indicates the first Information entropy of each factor; These are constant coefficients used to ensure information entropy. The range of values ​​is within interval; Based on information entropy, calculate the first... The redundancy (difference coefficient) of each factor is shown in the following formula:

[0042] in, Indicates the first Redundancy of each factor; Calculate the redundancy based on the number of... The objective weights of each factor are shown in the following formula:

[0043] in, Indicates the first The objective weights of each factor are calculated through iteration to obtain the objective weight vector. .

[0044] The multiplicative normalization method is used to fuse subjective and objective weights and objective weights to obtain the first... The intrinsic weights of each factor are shown in the following formula:

[0045] in, Indicates the first The intrinsic weights of each factor; Indicates the first Subjective weights of each factor; To sum the index variables, iterate from 1 to... , used to represent the index of all influencing factors; the intrinsic weight vector is obtained by iterating through the data. .

[0046] Step 3: Based on the real-time rainfall process parameters, the intrinsic weights are adjusted using a dynamic correction function to obtain the dynamic weights of each influencing parameter; wherein, the dynamic correction function takes rainfall intensity or cumulative rainfall as input and weight adjustment coefficient as output, and is used to reflect the time-varying enhancement or weakening effect of the rainfall process on the sensitivity of the source conditions and topographic features.

[0047] In some embodiments, the specific process of adjusting the intrinsic weights using a dynamic correction function based on real-time rainfall process parameters to obtain the dynamic weights of each influencing parameter is as follows: The dynamic correction function is defined as a two-parameter piecewise function with time-period rainfall intensity and effective cumulative rainfall as inputs. The dynamic correction function includes a source sensitivity correction sub-function and a topography sensitivity correction sub-function. The source sensitivity correction sub-function is used to characterize the time-varying enhancement or weakening effect of the rainfall process on the intrinsic weights corresponding to the source condition parameters, and the topography sensitivity correction sub-function is used to characterize the time-varying enhancement or weakening effect of the rainfall process on the intrinsic weights corresponding to the topography feature parameters. The real-time collected rainfall intensity and effective cumulative rainfall are simultaneously input into the source sensitivity correction sub-function and the terrain sensitivity correction sub-function, and the weight adjustment coefficients corresponding to each source condition parameter and each terrain feature parameter are calculated respectively. The intrinsic weights of each influencing parameter in the material source condition parameters and the intrinsic weights of each influencing parameter in the terrain feature parameters are multiplied by their corresponding weight adjustment coefficients to obtain the dynamic weights of each material source condition parameter and the dynamic weights of each terrain feature parameter. The dynamic weight vector of multiple influencing parameters is composed of the dynamic weights of the various source condition parameters and the dynamic weights of the various topographic feature parameters.

[0048] In some embodiments, both the source sensitivity correction subfunction and the terrain sensitivity correction subfunction adopt a piecewise nonlinear form. When the effective cumulative rainfall is lower than or equal to the lower limit of the first threshold interval, the weight adjustment coefficient increases linearly with the increase of rainfall intensity over time. When the effective cumulative rainfall is higher than the upper limit of the first threshold interval and lower than or equal to the lower limit of the second threshold interval, the weight adjustment coefficient increases exponentially with the increase of rainfall intensity over time. When the effective cumulative rainfall exceeds the upper limit of the second threshold interval, the weight adjustment coefficient reaches the saturation upper limit and no longer changes with the increase of rainfall intensity over time; The values ​​of the first threshold interval and the second threshold interval are determined based on the rainfall process data of historical debris flow events in the target watershed, as well as the sensitivity analysis results of the corresponding source conditions and topographic features, and are simultaneously reviewed and corrected when the intrinsic weights of the target watershed are updated.

[0049] The source sensitivity correction sub-function can be expressed by the following formula:

[0050] in, The source sensitivity correction function represents the rainfall intensity during the time period. and effective cumulative rainfall The adjustment coefficient (output value) of the intrinsic weights of the source condition parameters under the combined effect; and These are the first and second thresholds for effective cumulative rainfall, respectively. For source sensitivity in low cumulative rainfall range ( The linear growth coefficient of ) is used to control the weight adjustment coefficient with respect to rainfall intensity over time. The linear growth rate; For the source sensitivity in the medium cumulative rainfall range ( The exponential growth rate coefficient is used to control the overall amplification ratio of the exponential adjustment portion; The exponential growth rate coefficient of source sensitivity in the medium cumulative rainfall range; This is the reference rainfall intensity offset for source sensitivity within the cumulative rainfall range, used to adjust the exponential function. The starting point ensures that the function is continuous or conforms to the physical transition law at the junction of segments; For source sensitivity in high cumulative rainfall range ( The upper limit of saturation is a constant value.

[0051] The terrain sensitivity correction sub-function can be expressed by the following formula:

[0052] in, The terrain sensitivity correction sub-function represents the rainfall intensity during the time period. and effective cumulative rainfall The adjustment coefficient (output value) of the intrinsic weights corresponding to the terrain feature parameters under the combined effect. For terrain sensitivity in low cumulative rainfall range ( The linear growth coefficient of ) is used to control the weight adjustment coefficient with respect to rainfall intensity over time. The linear growth rate; For terrain sensitivity in the medium cumulative rainfall range ( The exponential growth rate coefficient is used to control the overall amplification ratio of the exponential adjustment portion; This represents the exponential growth rate coefficient of topographic sensitivity in the medium cumulative rainfall range; This is the reference rainfall intensity offset for topographic sensitivity over the cumulative rainfall range, used to adjust the exponential function. The starting point ensures that the function is continuous or conforms to the physical transition law at the junction of segments; For terrain sensitivity in high cumulative rainfall ranges ( The upper limit of saturation is a constant value.

[0053] Let the first The intrinsic weights of each source condition parameter are: , No. The eigenweight of each terrain feature parameter is The corresponding dynamic weights are:

[0054]

[0055] The final multi-influence parameter dynamic weight vector As shown below:

[0056] in, The number of source condition parameters. This represents the number of terrain feature parameters.

[0057] Step 4: Establish a critical function that couples dynamic weights with the critical condition for debris flow initiation. Substitute the weighted set of multiple influence parameters into the critical function to calculate the critical index for debris flow initiation under the current working condition.

[0058] In some embodiments, the specific process of establishing the critical function that couples the dynamic weights with the critical condition for debris flow initiation is as follows: Based on the physical mechanism of debris flow initiation, the basic form of the critical function is a ratio function of the weighted driving term and the critical resistance term. The weighted driving term is used to characterize the comprehensive driving capacity of the material source conditions and terrain features on debris flow initiation under dynamic weighting. The critical resistance term is used to characterize the critical resistance conditions that need to be overcome for debris flow initiation under the current working conditions. The source condition parameters and terrain feature parameters in the multi-influence parameter set are multiplied by their corresponding dynamic weights to obtain a weighted source parameter vector and a weighted terrain parameter vector, respectively. The weighted source parameter vector and the weighted terrain parameter vector are then summed element by element to obtain a comprehensive driving index. Based on the source conditions and topographic features of the target watershed, a critical resistance term is constructed using the limit equilibrium analysis method; wherein, the critical resistance term includes the resistance component provided by the shear strength of the source and the resistance component provided by the channel geometric constraints; The ratio of the comprehensive driving index to the critical resistance term is defined as the debris flow initiation critical exponent, and the critical function is established as shown in the following formula:

[0059] in, The critical index for the initiation of debris flows; For the first The weight values ​​of each influencing parameter under dynamic weighting; For the first Normalized values ​​of the influencing parameters; The number of influencing parameters; For the first One resistance component; This represents the quantity of the resistance component.

[0060] Step 5: When the debris flow initiation critical index exceeds the preset threshold, a debris flow initiation warning message is issued.

[0061] In some embodiments, the specific process of issuing a debris flow initiation warning message when the debris flow initiation critical index exceeds a preset threshold is as follows: Based on the limit equilibrium critical conditions for debris flow initiation in the target watershed, measured samples of critical indices corresponding to historical disaster events, and statistical samples of critical indices for historical conditions without disasters, a multi-level progressive early warning threshold sequence corresponding to the risk level of debris flow initiation is calibrated. The multi-level progressive early warning threshold sequence is divided into three threshold levels based on the critical equilibrium state of the driving and resistance terms corresponding to the critical function: low-risk early warning threshold, medium-risk early warning threshold, and high-risk early warning threshold. The values ​​of each threshold level are synchronously reviewed and corrected as the intrinsic weights of the impact parameters of the target watershed are updated. According to the time step consistent with the real-time rainfall process parameters, the debris flow initiation critical index is updated and calculated synchronously. The debris flow initiation critical index calculated in real time is compared with the multi-level progressive early warning threshold sequence level by level to determine the debris flow risk level corresponding to the current working condition. When the real-time calculated debris flow initiation critical index exceeds the warning threshold of the corresponding level, a threshold exceedance persistence check is initiated. The threshold exceedance persistence check involves continuously checking the debris flow initiation critical index for no less than three preset time steps. If the debris flow initiation critical index is continuously higher than the warning threshold of the corresponding level throughout the entire check period, the warning triggering condition of the corresponding level is determined to be met. If the debris flow initiation critical index falls below the corresponding warning threshold during the check period, the check is terminated and the warning of the corresponding level is not triggered. Once the warning triggering conditions are met, based on the dynamic weight vector of each influencing parameter under the current working conditions, the contribution ratio of each influencing parameter to the debris flow initiation critical index is calculated, the top two dominant influencing parameters with the highest contribution ratio are identified, and warning information including debris flow risk level, dominant influencing parameters, real-time rainfall conditions, and the time series change trend of the critical index is generated. The warning information is released according to the preset warning release level and release rules. The calculation and update time step of the debris flow initiation critical index is compressed to half of the original step step, and encrypted calculation and dynamic tracking are performed until the debris flow initiation critical index falls back below the low-risk warning threshold.

[0062] Based on the same inventive concept, and corresponding to any of the above embodiments, refer to... Figure 2 This invention provides a debris flow initiation critical early warning system based on dynamic weights of multiple influencing parameters, used to implement the aforementioned debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters, comprising: The parameter acquisition module is configured to acquire source condition parameters, terrain feature parameters, and real-time rainfall process parameters through field surveys, remote sensing interpretation, digital elevation models, field mapping, and rainfall monitoring networks deployed in the target watershed. The acquired parameters are then parameterized according to a unified time scale and spatial resolution to construct a multi-influence parameter set containing source condition feature vectors, terrain feature vectors, and rainfall process feature vectors. The intrinsic weight calculation module, whose input end is connected to the output end of the parameter acquisition module, is configured as follows: based on the physical mechanism of debris flow initiation, an intrinsic weight function reflecting the contribution of source conditions and terrain features to debris flow initiation is constructed; according to the source condition parameters and terrain feature parameters output by the parameter acquisition module, the intrinsic weight of each influencing parameter is calculated by combining the analytic hierarchy process and the entropy weight method, and the intrinsic weight vector is output. A dynamic weight correction module, whose input is connected to the output of the parameter acquisition module and the output of the intrinsic weight calculation module, is configured to: calculate the weight adjustment coefficients corresponding to each source condition parameter and each terrain feature parameter respectively using a dynamic correction function that includes a source sensitivity correction sub-function and a terrain sensitivity correction sub-function, based on the real-time rainfall process parameters output by the parameter acquisition module; dynamically adjust the intrinsic weight vector output by the intrinsic weight calculation module using the weight adjustment coefficients; and output a dynamic weight vector of multiple influencing parameters. The critical index calculation module, whose input is connected to the output of the parameter acquisition module and the output of the dynamic weight correction module, is configured to: construct a critical function that couples dynamic weights with the critical conditions for debris flow initiation based on the physical mechanism of debris flow initiation; substitute the set of multiple influence parameters output by the parameter acquisition module and the dynamic weight vector output by the dynamic weight correction module into the critical function; calculate the critical index for debris flow initiation under the current working condition; and output the critical index for debris flow initiation. The early warning release module, whose input is connected to the output of the critical index calculation module, is configured to: compare the debris flow initiation critical index output by the critical index calculation module with a preset multi-level progressive early warning threshold sequence; and generate and release early warning information including debris flow risk level, dominant influencing parameters, real-time rainfall conditions, and the time-series change trend of the critical index when the debris flow initiation critical index is continuously higher than the corresponding level's early warning threshold for at least three consecutive preset time steps.

[0063] Based on the same inventive concept, corresponding to any of the above embodiments, the present invention provides an electronic device, including a memory and a processor. The memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the debris flow initiation critical early warning method based on dynamic weights of multiple influence parameters according to the embodiments.

[0064] Alternatively, the aforementioned electronic device may be a server.

[0065] In addition, this embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the debris flow initiation critical early warning method based on dynamic weights of multiple influence parameters of the embodiment.

[0066] It is understood that the processor in the embodiments of the present invention may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0067] The method steps in the embodiments of the present invention can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an ASIC.

[0068] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a storage medium or transmitted through a storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive (SSD)).

Claims

1. A critical early warning method for debris flow initiation based on dynamic weights of multiple influencing parameters, characterized in that, Includes the following steps: Obtain source condition parameters, topographic feature parameters, and real-time rainfall process parameters of the target watershed, and construct a set of multiple influence parameters; Based on the physical mechanism of debris flow initiation, an intrinsic weight function reflecting the contribution of source conditions and topographic features to debris flow initiation is constructed, and the intrinsic weights of each influencing parameter are calculated according to the source condition parameters and topographic feature parameters. Based on real-time rainfall process parameters, the intrinsic weights are adjusted using a dynamic correction function to obtain the dynamic weights of each influencing parameter; wherein, the dynamic correction function takes rainfall intensity or cumulative rainfall as input and weight adjustment coefficient as output, and is used to reflect the time-varying enhancement or weakening effect of the rainfall process on the sensitivity of the source conditions and topographic features; Establish a critical function that couples dynamic weights with the critical condition for debris flow initiation, substitute the weighted set of multiple influence parameters into the critical function, and calculate the critical index for debris flow initiation under the current working condition. When the debris flow initiation critical index exceeds a preset threshold, a debris flow initiation warning message is issued.

2. The debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters as described in claim 1, characterized in that... The specific process of obtaining the source condition parameters, topographic feature parameters, and real-time rainfall process parameters of the target watershed, and constructing a multi-influence parameter set is as follows: The source conditions parameters of the target watershed are obtained through on-site investigation and remote sensing interpretation. These source conditions parameters include the reserves of solid materials that can participate in the initiation of debris flows, the material gradation characteristics, and the internal friction angle of the materials. Topographic feature parameters are obtained through digital elevation modeling and field surveying. These parameters include the longitudinal gradient of the gully bed, the catchment area of ​​the watershed, and the morphological ratio of the gully width to its depth. Rainfall process parameters are collected in real time through a rainfall monitoring network deployed in the target watershed. These parameters include rainfall intensity, effective cumulative rainfall, and rainfall duration over a given period. The parameters of the material source conditions, the parameters of the terrain features, and the parameters of the real-time rainfall process are parameterized according to a unified time scale and spatial resolution, forming a multi-influence parameter set that includes the feature vectors of the material source conditions, the feature vectors of the terrain, and the feature vectors of the rainfall process.

3. The debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters as described in claim 2, characterized in that, The intrinsic weighting function, which reflects the contribution of source conditions and topographic features to debris flow initiation based on the physical mechanism of debris flow, is constructed. The specific process of calculating the intrinsic weights of each influencing parameter based on the source condition parameters and topographic feature parameters is as follows: The solid material reserves, material gradation characteristics, and material internal friction angle in the material source condition parameters are parameterized as material source supply factor, gradation characteristic factor, and shear strength factor, respectively; the gully bed longitudinal gradient, watershed catchment area, and gully morphology ratio in the topographic characteristic parameters are parameterized as topographic driving factor, runoff capacity factor, and gully constraint factor, respectively. Based on the physical coupling mechanism of sediment supply and topographic driving during debris flow initiation, an intrinsic weight function is constructed with sediment supply factor, gradation characteristic factor, shear strength factor, topographic driving factor, runoff capacity factor, and channel constraint factor as independent variables. The intrinsic weight function uses a combination of the analytic hierarchy process (AHP) and the entropy weight method to determine the inherent contribution of each factor in debris flow initiation. Specifically, the AHP constructs a judgment matrix based on the physical mechanism of debris flow initiation, which is used to determine the subjective weight of each factor. The entropy weight method calculates the objective weight of each factor based on historical sediment supply conditions and topographic feature data of the target watershed. The intrinsic weight of each factor is obtained by multiplying and normalizing the subjective and objective weights. The intrinsic weight vector, which characterizes the inherent contribution of each influencing parameter, is obtained by multiplying the material supply factor, gradation characteristic factor, shear strength factor, terrain driving factor, runoff capacity factor, and channel constraint factor by their respective intrinsic weights.

4. The debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters as described in claim 1, characterized in that, The specific process of adjusting the intrinsic weights based on real-time rainfall parameters using a dynamic correction function to obtain the dynamic weights of each influencing parameter is as follows: The dynamic correction function is defined as a two-parameter piecewise function with time-period rainfall intensity and effective cumulative rainfall as inputs. The dynamic correction function includes a source sensitivity correction sub-function and a topography sensitivity correction sub-function. The source sensitivity correction sub-function is used to characterize the time-varying enhancement or weakening effect of the rainfall process on the intrinsic weights corresponding to the source condition parameters, and the topography sensitivity correction sub-function is used to characterize the time-varying enhancement or weakening effect of the rainfall process on the intrinsic weights corresponding to the topography feature parameters. The real-time collected rainfall intensity and effective cumulative rainfall are simultaneously input into the source sensitivity correction sub-function and the terrain sensitivity correction sub-function, and the weight adjustment coefficients corresponding to each source condition parameter and each terrain feature parameter are calculated respectively. The intrinsic weights of each influencing parameter in the material source condition parameters and the intrinsic weights of each influencing parameter in the terrain feature parameters are multiplied by their corresponding weight adjustment coefficients to obtain the dynamic weights of each material source condition parameter and the dynamic weights of each terrain feature parameter. The dynamic weight vector of multiple influencing parameters is composed of the dynamic weights of the various source condition parameters and the dynamic weights of the various topographic feature parameters.

5. The debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters as described in claim 4, characterized in that, Both the source sensitivity correction sub-function and the terrain sensitivity correction sub-function adopt a piecewise nonlinear form. When the effective cumulative rainfall is lower than or equal to the lower limit of the first threshold interval, the weight adjustment coefficient increases linearly with the increase of rainfall intensity over time. When the effective cumulative rainfall is higher than the upper limit of the first threshold interval and lower than or equal to the lower limit of the second threshold interval, the weight adjustment coefficient increases exponentially with the increase of rainfall intensity over time. When the effective cumulative rainfall exceeds the upper limit of the second threshold interval, the weight adjustment coefficient reaches the saturation upper limit and no longer changes with the increase of rainfall intensity over time; The values ​​of the first threshold interval and the second threshold interval are determined based on the rainfall process data of historical debris flow events in the target watershed, as well as the sensitivity analysis results of the corresponding source conditions and topographic features, and are simultaneously reviewed and corrected when the intrinsic weights of the target watershed are updated.

6. The debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters as described in claim 1, characterized in that, The specific process for establishing the critical function that couples the dynamic weights with the critical condition for debris flow initiation is as follows: Based on the physical mechanism of debris flow initiation, the basic form of the critical function is a ratio function of the weighted driving term and the critical resistance term. The weighted driving term is used to characterize the comprehensive driving capacity of the material source conditions and terrain features on debris flow initiation under dynamic weighting. The critical resistance term is used to characterize the critical resistance conditions that need to be overcome for debris flow initiation under the current working conditions. The source condition parameters and terrain feature parameters in the multi-influence parameter set are multiplied by their corresponding dynamic weights to obtain a weighted source parameter vector and a weighted terrain parameter vector, respectively. The weighted source parameter vector and the weighted terrain parameter vector are then summed element by element to obtain a comprehensive driving index. Based on the source conditions and topographic features of the target watershed, a critical resistance term is constructed using the limit equilibrium analysis method; wherein, the critical resistance term includes the resistance component provided by the shear strength of the source and the resistance component provided by the channel geometric constraints; The ratio of the comprehensive driving index to the critical resistance term is defined as the debris flow initiation critical exponent, and the critical function is established as shown in the following formula: in, The critical index for the initiation of debris flows; For the first The weight values ​​of each influencing parameter under dynamic weighting; For the first Normalized values ​​of the influencing parameters; The number of influencing parameters; For the first One resistance component; This represents the quantity of the resistance component.

7. The debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters as described in claim 6, characterized in that, The specific process for issuing a debris flow initiation warning when the debris flow initiation critical index exceeds a preset threshold is as follows: Based on the limit equilibrium critical conditions for debris flow initiation in the target watershed, measured samples of critical indices corresponding to historical disaster events, and statistical samples of critical indices for historical conditions without disasters, a multi-level progressive early warning threshold sequence corresponding to the risk level of debris flow initiation is calibrated. The multi-level progressive early warning threshold sequence is divided into three threshold levels based on the critical equilibrium state of the driving and resistance terms corresponding to the critical function: low-risk early warning threshold, medium-risk early warning threshold, and high-risk early warning threshold. The values ​​of each threshold level are synchronously reviewed and corrected as the intrinsic weights of the impact parameters of the target watershed are updated. According to the time step consistent with the real-time rainfall process parameters, the debris flow initiation critical index is updated and calculated synchronously. The debris flow initiation critical index calculated in real time is compared with the multi-level progressive early warning threshold sequence level by level to determine the debris flow risk level corresponding to the current working condition. When the real-time calculated debris flow initiation critical index exceeds the warning threshold of the corresponding level, a threshold exceedance persistence check is initiated. The threshold exceedance persistence check involves continuously checking the debris flow initiation critical index for no less than three preset time steps. If the debris flow initiation critical index is continuously higher than the warning threshold of the corresponding level throughout the entire check period, the warning triggering condition of the corresponding level is determined to be met. If the debris flow initiation critical index falls below the corresponding warning threshold during the check period, the check is terminated and the warning of the corresponding level is not triggered. Once the warning triggering conditions are met, based on the dynamic weight vector of each influencing parameter under the current working conditions, the contribution ratio of each influencing parameter to the debris flow initiation critical index is calculated, the top two dominant influencing parameters with the highest contribution ratio are identified, and warning information including debris flow risk level, dominant influencing parameters, real-time rainfall conditions, and the time series change trend of the critical index is generated. The warning information is released according to the preset warning release level and release rules. The calculation and update time step of the debris flow initiation critical index is compressed to half of the original step step, and encrypted calculation and dynamic tracking are performed until the debris flow initiation critical index falls back below the low-risk warning threshold.

8. A debris flow initiation critical early warning system based on dynamic weights of multiple influencing parameters, used to implement the debris flow initiation critical early warning method based on dynamic weights of multiple influencing parameters as described in any one of claims 1-7, characterized in that, include: The parameter acquisition module is configured to acquire source condition parameters, terrain feature parameters, and real-time rainfall process parameters through field surveys, remote sensing interpretation, digital elevation models, field mapping, and rainfall monitoring networks deployed in the target watershed. The acquired parameters are then parameterized according to a unified time scale and spatial resolution to construct a multi-influence parameter set containing source condition feature vectors, terrain feature vectors, and rainfall process feature vectors. The intrinsic weight calculation module, whose input end is connected to the output end of the parameter acquisition module, is configured as follows: based on the physical mechanism of debris flow initiation, an intrinsic weight function reflecting the contribution of source conditions and terrain features to debris flow initiation is constructed; according to the source condition parameters and terrain feature parameters output by the parameter acquisition module, the intrinsic weight of each influencing parameter is calculated by combining the analytic hierarchy process and the entropy weight method, and the intrinsic weight vector is output. A dynamic weight correction module, whose input is connected to the output of the parameter acquisition module and the output of the intrinsic weight calculation module, is configured to: calculate the weight adjustment coefficients corresponding to each source condition parameter and each terrain feature parameter respectively using a dynamic correction function that includes a source sensitivity correction sub-function and a terrain sensitivity correction sub-function, based on the real-time rainfall process parameters output by the parameter acquisition module; dynamically adjust the intrinsic weight vector output by the intrinsic weight calculation module using the weight adjustment coefficients; and output a dynamic weight vector of multiple influencing parameters. The critical index calculation module, whose input is connected to the output of the parameter acquisition module and the output of the dynamic weight correction module, is configured to: construct a critical function that couples dynamic weights with the critical conditions for debris flow initiation based on the physical mechanism of debris flow initiation; substitute the set of multiple influence parameters output by the parameter acquisition module and the dynamic weight vector output by the dynamic weight correction module into the critical function; calculate the critical index for debris flow initiation under the current working condition; and output the critical index for debris flow initiation. The early warning release module, whose input is connected to the output of the critical index calculation module, is configured to: compare the debris flow initiation critical index output by the critical index calculation module with a preset multi-level progressive early warning threshold sequence; and generate and release early warning information including debris flow risk level, dominant influencing parameters, real-time rainfall conditions, and the time-series change trend of the critical index when the debris flow initiation critical index is continuously higher than the corresponding level's early warning threshold for at least three consecutive preset time steps.

9. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program and the processor runs the computer program to enable the electronic device to perform the debris flow initiation critical early warning method based on dynamic weights of multiple influence parameters as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the debris flow initiation critical early warning method based on dynamic weights of multiple influence parameters as described in any one of claims 1-7.

Images