Flood risk assessment and damage calculation method based on tail risk state evolution

Abstract

The application discloses a flood risk assessment and damage calculation method based on tail risk state evolution, and belongs to the field of extreme flood risk assessment and damage calculation. The method comprises the following steps: collecting time-labeled historical observation data of historical flood events in a target area; obtaining a flood risk data set; initializing a tail risk characteristic function by using the flood risk data set; performing time scale conversion on the tail risk characteristic function, and predicting the occurrence frequency of each flood intensity index level in a future target period by using the converted tail risk characteristic function; determining the damage rate of each predicted flood scenario; multiplying the occurrence frequency, damage rate and exposure base of each predicted flood scenario corresponding to the flood intensity index level to obtain the future damage scale corresponding to each predicted flood scenario; and obtaining the total flood disaster risk impact damage amount. The application improves the risk identification precision and impact assessment reliability of extreme flood events under non-stationary conditions.

Description

Technical Field

[0001] This invention belongs to the field of extreme flood risk assessment and damage calculation, and in particular relates to a flood risk assessment and damage calculation method based on the evolution of tail risk state. Background Technology

[0002] In recent years, under the combined influence of global climate change and rapid urbanization, the frequency and intensity of extreme rainfall events have been increasing, leading to more significant suddenness and complexity in river floods, flash floods, and urban flooding. Flood disasters not only cause widespread infrastructure damage and economic losses, but also exhibit clear spatial diffusion characteristics, easily transmitting risks to adjacent areas through river networks, topography, and urban drainage networks, thus continuously expanding the scope of flood impact. Therefore, accurately identifying the evolution process of extreme flood risks and conducting damage assessments has become an important research direction in the field of flood risk management.

[0003] Currently, flood risk assessment research mainly focuses on flood probability analysis, flood loss estimation, and spatial distribution simulation of disasters. Traditional methods typically rely on historical flood observation data, using indicators such as rainfall, water level, flow rate, or inundated area to describe flood intensity. They then analyze the likelihood of flood occurrence using probabilistic statistical models and combine this with empirical loss rate models to estimate the scale of disaster damage. While these methods have some applicability under normal flood conditions, they still have significant shortcomings in identifying extreme flood risks and analyzing cross-regional risk propagation.

[0004] Existing flood risk assessment and damage measurement methods suffer from the following main shortcomings: First, existing methods often use overall historical samples to construct risk distribution models, typically fitting flood intensity indicators based on a single probability distribution function, making it difficult to accurately identify the statistical characteristics of extreme flood events in their tail regions. Because extreme flood events are characterized by low frequency and large damage scale, their tail samples often exhibit significant non-stationarity and phased changes. Traditional fixed-parameter models struggle to reflect the dynamic evolution of tail risk states over time, resulting in low accuracy in predicting the frequency of extreme floods.

[0005] Secondly, existing methods in flood risk analysis typically focus only on risk changes within a single spatial unit, failing to adequately consider the propagation and correlation of floods between different regions. Flood disasters are significantly spatially dependent, influenced by topographic relief, river connectivity, and regional drainage conditions. A flood in one region can affect the risk levels of adjacent regions through spatial transmission. Most existing risk models lack descriptions of inter-regional spatial propagation mechanisms, making it difficult to effectively characterize the interconnected diffusion process of floods at a spatial scale, thus limiting the ability to assess overall regional risk.

[0006] Furthermore, existing damage assessment methods typically estimate disaster damage scale based on fixed damage rates or empirical loss functions, lacking a comprehensive consideration of the spatial exposure characteristics, vulnerability differences, and regional propagation effects of affected bodies. When flood intensity exceeds the threshold for local complete inundation, traditional models often directly adopt the fixed total loss assumption, ignoring the repair capacity of affected bodies and the risk superposition effect caused by spatial diffusion, which can easily lead to biased damage estimation. Meanwhile, existing research often treats flood tail risk identification, spatial propagation analysis, and damage assessment processes separately, lacking a comprehensive method that can couple the evolution of tail risk status with spatial propagation mechanisms, making it difficult to meet the practical needs of risk assessment and damage prediction under extreme flood conditions. Summary of the Invention

[0007] To address the aforementioned shortcomings in existing technologies, this invention provides a flood risk assessment and damage calculation method based on tail risk state evolution, which improves the accuracy of risk identification and the reliability of impact assessment for extreme flood events under non-stationary conditions.

[0008] To achieve the aforementioned objectives, the present invention employs the following technical solution: a flood risk assessment and damage calculation method based on the evolution of tail risk states, comprising: Collect time-stamped historical observation data of historical flood events in the target area; the historical observation data includes flood intensity indicators, loss data, and basic information data of the disaster-affected areas. The flood intensity index of historical flood events is discretized using the flood intensity index classification threshold to obtain a flood risk dataset; Initialize the tail risk feature function using a flood risk dataset; set a sliding observation window, and during the flood evolution, update the tail risk feature function based on the observation data of newly added flood events within the current sliding observation window when the sliding observation window is updated; Based on the timescale relationship between the historical observation time length and the predicted time length of the future target time period, the tail risk characteristic function is transformed by timescale, and the frequency of occurrence of each flood intensity index level is predicted within the future target time period using the transformed tail risk characteristic function. Based on the exposure base, loss data and flood intensity index level of disaster-bearing bodies under various historical flood events, a nonlinear growth function is fitted to obtain the dynamic response relationship between flood intensity level and damage rate. The propagation process of floods under various flood intensity index levels is simulated using a flood evolution model to obtain the inundation range of the predicted flood scenario corresponding to each flood intensity index level. Based on the inundation range of each predicted flood scenario, the exposure base of the disaster-bearing bodies within the inundation range is summed to obtain the exposure base of the disaster-bearing bodies for each predicted flood scenario; The damage rate for each predicted flood scenario is determined by utilizing the dynamic response relationship between flood intensity level and damage rate. Multiply the frequency of occurrence, damage rate and exposure base of flood intensity index level corresponding to each predicted flood scenario by the product of the two factors to obtain the future damage scale corresponding to each predicted flood scenario. The total damage caused by the overall flood risk is obtained by summing the future damage scales corresponding to each predicted flood scenario.

[0009] Furthermore, the flood intensity index is flood flow or flood velocity; the basic information data of the disaster-affected area includes the digital elevation model of the disaster-affected area, river geometry data, land use data, and building data.

[0010] Furthermore, the discretization specifically involves discretizing the flood intensity index of historical flood events into the lower limit value of the flood intensity index level of the historical flood events; the flood risk dataset... The expression is:

[0011] in, For the first The lower limit of each flood intensity index level; To statistically analyze flood intensity indices based on discretized historical flood events. The frequency of occurrence.

[0012] Furthermore, the initialization of the tail risk feature function is specifically as follows: A set of distribution functions is constructed to fit the probability distribution law of flood intensity index, and the set of distribution functions includes several candidate probability distribution function models; Each candidate probability distribution function model was fitted using the flood risk tail distribution dataset; The fitting results of each candidate probability distribution function model are evaluated based on the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC), and the optimal probability distribution function model is selected. The optimal probability distribution function model obtained through screening is used to fit the tail distribution dataset of flood risk, and a tail risk feature function is constructed.

[0013] Furthermore, the tail risk feature function is updated based on the observation data of newly added flood events within the current sliding observation window, specifically as follows: Calculate the relative error between the observed data of newly added flood events within the sliding observation window and the current tail risk feature function prediction results:

[0014] in, The relative error between newly observed flood event data and the prediction results of the tail risk characteristic function; When a new flood event occurs, the discrete value of the corresponding flood intensity index within the sliding observation window. The actual frequency of observations; Discrete values ​​of flood intensity index calculated based on the current tail risk characteristic function. The frequency of predictions; It is a set of discrete values ​​for flood intensity indices; The total number of elements in the discrete set of flood intensity index values; for The absolute value; Determine if the relative error is less than 10%. If yes, no action is taken; otherwise, introduce tail evolution weight coefficients to update the tail risk feature function.

[0015]

[0016] in, This is the updated tail risk feature function; The tail risk feature function before the update; Tail evolution weighting coefficients are used to characterize the change in the current flood sequence volatility relative to the historical overall volatility. This represents the standard deviation of the discrete values ​​of the flood intensity index for each flood event within the current sliding observation window; This represents the standard deviation of the discrete values ​​of flood intensity indices for each flood event within the total historical time span. This represents the discrete value of the flood intensity index.

[0017] Furthermore, the expression for the frequency of occurrence of each flood intensity index level within the target future time period is as follows:

[0018]

[0019] in, This represents the overall frequency of flood events within the target future timeframe. This is a dynamic time correction factor; This is the current tail risk characteristic function; This is the start time of the target time period; This is the end time of the target time period; This represents the number of dates that occurred within the current sliding observation window; This represents the number of dates within the timeframe of all historical observations that coincided with flood events. This represents the discrete value of the flood intensity index.

[0020] Furthermore, the dynamic response relationship between flood intensity level and damage rate is obtained as follows: The ratio of the exposure base and loss data of disaster-bearing bodies under each historical flood event is used as the physical damage rate under each historical flood event. Based on the flood intensity index level of each historical flood event, the physical damage rate and the flood intensity index level are correlated to obtain the damage rate-intensity level response relationship dataset. Based on the damage rate-intensity level response relationship dataset, a nonlinear growth function is fitted to obtain the dynamic response relationship between flood intensity level and damage rate.

[0021] Furthermore, the expression for the dynamic response relationship between the flood intensity level and the damage rate is as follows:

[0022] in, For the first Damage rate corresponding to each flood intensity level; , and All are fitted parameters; For the first Discrete values ​​corresponding to each flood intensity index level.

[0023] Furthermore, the determination of the damage rate for each predicted flood scenario specifically involves: when the flood intensity index corresponding to the predicted flood scenario does not reach the complete inundation threshold, the damage rate is determined based on the dynamic response relationship between the flood intensity level and the damage rate; when the flood intensity index corresponding to the predicted flood scenario reaches the preset flood intensity threshold corresponding to the complete inundation of the disaster-bearing body, the damage rate is determined based on the repairability of the disaster-bearing body.

[0024] Furthermore, the repairability of the disaster-bearing body is determined based on the post-disaster recovery capability index corresponding to the type of disaster-bearing body; the post-disaster recovery capability index includes at least one of the following: recovery time index, recovery cost index, functional recovery index, and resource guarantee index; the relationship between the repairability of the disaster-bearing body and the damage rate is established based on historical disaster recovery samples, including the following steps: Collect historical disaster recovery samples, which include indicators of the post-disaster recovery capacity of disaster-bearing bodies and historical damage rates under the highest upper limit of the flood intensity index level; The post-disaster recovery capacity indicators for each historical disaster recovery sample were standardized, and the repairability index of the disaster-bearing body corresponding to each historical disaster recovery sample was calculated separately:

[0025] in, The index represents the repairability of the disaster-bearing body. The quantity of indicators for post-disaster recovery capacity; For the first Standardized values ​​for post-disaster recovery capacity indicators; For the first The weights corresponding to various disaster recovery capability indicators; Based on the repairability index of the disaster-bearing body corresponding to each historical disaster recovery sample, a damage rate mapping relationship is established:

[0026] in, The damage rate of the disaster-bearing body after it reaches the threshold of complete flooding; This is a mapping function established through statistical analysis or regression fitting.

[0027] The beneficial effects of this invention are as follows: By constructing a tail risk feature function, this invention fits the tail data of extreme flood events and updates the tail risk status in conjunction with a time window, thereby improving the model's ability to identify the low-frequency, high-damage characteristics of extreme floods and enhancing its ability to characterize the evolution of flood risk under non-stationary conditions. An adaptive update mechanism based on the dynamic changes in tail risk is proposed. By combining the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) to select the optimal probability distribution function model, and by performing segmented processing of the tail based on the stage-specific changes in historical observation data, the tail risk feature function is dynamically corrected, improving the accuracy of predicting the frequency of extreme flood events over different time spans in the future and enhancing the model's applicability to extrapolation analysis of extreme event risks. Furthermore, a conditional damage response mechanism for disaster-bearing bodies is constructed. By establishing a damage rate-intensity index response function and combining a layered damage processing method, damage characteristics in incompletely submerged and completely submerged states are differentiated and calculated, achieving dynamic adjustment of the damage rate as the flood intensity changes, thus improving the rationality and accuracy of future flood damage scale calculations. By coupling tail risk state evolution analysis, spatial propagation mechanism and damage calculation process, it is possible to comprehensively predict the total damage caused by future flood risks based on the frequency of extreme floods, regional risk diffusion and differences in the vulnerability of disaster-bearing bodies, thus providing more scientific technical support for extreme flood risk management, loss prediction and disaster prevention and mitigation decision-making. Attached Figure Description

[0028] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0029] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0030] like Figure 1 As shown, in one embodiment of the present invention, a flood risk assessment and damage calculation method based on tail risk state evolution includes: Collect time-stamped historical observation data of historical flood events in the target area; the historical observation data includes flood intensity indicators, loss data, and basic information data of the disaster-affected areas. The flood intensity index of historical flood events is discretized using the flood intensity index classification threshold to obtain a flood risk dataset; Initialize the tail risk feature function using a flood risk dataset; set a sliding observation window, and during the flood evolution, update the tail risk feature function based on the observation data of newly added flood events within the current sliding observation window when the sliding observation window is updated; Based on the timescale relationship between the historical observation time length and the predicted time length of the future target time period, the tail risk characteristic function is transformed by timescale, and the frequency of occurrence of each flood intensity index level is predicted within the future target time period using the transformed tail risk characteristic function. Based on the exposure base, loss data and flood intensity index level of disaster-bearing bodies under various historical flood events, a nonlinear growth function is fitted to obtain the dynamic response relationship between flood intensity level and damage rate. The propagation process of floods under various flood intensity index levels is simulated using a flood evolution model to obtain the inundation range of the predicted flood scenario corresponding to each flood intensity index level. Based on the inundation range of each predicted flood scenario, the exposure base of the disaster-bearing bodies within the inundation range is summed to obtain the exposure base of the disaster-bearing bodies for each predicted flood scenario; The damage rate for each predicted flood scenario is determined by utilizing the dynamic response relationship between flood intensity level and damage rate. Multiply the frequency of occurrence, damage rate and exposure base of flood intensity index level corresponding to each predicted flood scenario by the product of the two factors to obtain the future damage scale corresponding to each predicted flood scenario. The total damage caused by the overall flood risk is obtained by summing the future damage scales corresponding to each predicted flood scenario.

[0031] The flood intensity index is the flood flow rate or flood velocity; the basic information data of the disaster-affected area includes the digital elevation model of the disaster-affected area, river geometric data, land use data, and building data.

[0032] The discretization specifically involves discretizing the flood intensity index of historical flood events into the lower limit value of the flood intensity index level of the historical flood events; the flood risk dataset... The expression is:

[0033] in, For the first The lower limit of each flood intensity index level; To statistically analyze flood intensity indices based on discretized historical flood events. The frequency of occurrence.

[0034] The initialization of the tail risk feature function is specifically as follows: A set of distribution functions is constructed to fit the probability distribution law of flood intensity index, and the set of distribution functions includes several candidate probability distribution function models; Each candidate probability distribution function model was fitted using the flood risk tail distribution dataset; The fitting results of each candidate probability distribution function model are evaluated based on the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC), and the optimal probability distribution function model is selected. The optimal probability distribution function model obtained through screening is used to fit the tail distribution dataset of flood risk, and a tail risk feature function is constructed.

[0035] The tail risk feature function is updated based on the observation data of newly added flood events within the current sliding observation window, specifically as follows: Calculate the relative error between the observed data of newly added flood events within the sliding observation window and the current tail risk feature function prediction results:

[0036] in, The relative error between newly observed flood event data and the prediction results of the tail risk characteristic function; When a new flood event occurs, the discrete value of the corresponding flood intensity index within the sliding observation window. The actual frequency of observations; Discrete values ​​of flood intensity index calculated based on the current tail risk characteristic function. The frequency of predictions; It is a set of discrete values ​​for flood intensity indices; The total number of elements in the discrete set of flood intensity index values; for The absolute value; Determine if the relative error is less than 10%. If yes, no action is taken; otherwise, introduce tail evolution weight coefficients to update the tail risk feature function.

[0037]

[0038] in, This is the updated tail risk feature function; The tail risk feature function before the update; Tail evolution weighting coefficients are used to characterize the change in the current flood sequence volatility relative to the historical overall volatility. This represents the standard deviation of the discrete values ​​of the flood intensity index for each flood event within the current sliding observation window; This represents the standard deviation of the discrete values ​​of flood intensity indices for each flood event within the total historical time span. This represents the discrete value of the flood intensity index.

[0039] The expression for the frequency of occurrence of each flood intensity index level within the target future time period is as follows:

[0040]

[0041] in, This represents the overall frequency of flood events within the target future timeframe. This is a dynamic time correction factor; This is the current tail risk characteristic function; This is the start time of the target time period; This is the end time of the target time period; This represents the number of dates that occurred within the current sliding observation window; This represents the number of dates within the timeframe of all historical observations that coincided with flood events. This represents the discrete value of the flood intensity index.

[0042] The dynamic response relationship between flood intensity level and damage rate is obtained as follows: The ratio of the exposure base and loss data of disaster-bearing bodies under each historical flood event is used as the physical damage rate under each historical flood event. Based on the flood intensity index level of each historical flood event, the physical damage rate and the flood intensity index level are correlated to obtain the damage rate-intensity level response relationship dataset. Based on the damage rate-intensity level response relationship dataset, a nonlinear growth function is fitted to obtain the dynamic response relationship between flood intensity level and damage rate.

[0043] The expression for the dynamic response relationship between the flood intensity level and the damage rate is as follows:

[0044] in, For the first Damage rate corresponding to each flood intensity level; , and All are fitted parameters; For the first Discrete values ​​corresponding to each flood intensity index level.

[0045] The determination of the damage rate for each predicted flood scenario is specifically as follows: when the flood intensity index corresponding to the predicted flood scenario does not reach the complete inundation threshold, the damage rate is determined based on the dynamic response relationship between the flood intensity level and the damage rate; when the flood intensity index corresponding to the predicted flood scenario reaches the preset flood intensity threshold corresponding to the complete inundation of the disaster-bearing body, the damage rate is determined based on the repairability of the disaster-bearing body.

[0046] The repairability of the disaster-bearing body is determined based on the post-disaster recovery capability index corresponding to the type of disaster-bearing body; the post-disaster recovery capability index includes at least one of the following: recovery time index, recovery cost index, functional recovery index, and resource guarantee index; the relationship between the repairability of the disaster-bearing body and the damage rate is established based on historical disaster recovery samples, including the following steps: Collect historical disaster recovery samples, which include indicators of the post-disaster recovery capacity of disaster-bearing bodies and historical damage rates under the highest upper limit of the flood intensity index level; The post-disaster recovery capacity indicators for each historical disaster recovery sample were standardized, and the repairability index of the disaster-bearing body corresponding to each historical disaster recovery sample was calculated separately:

[0047] in, The index represents the repairability of the disaster-bearing body. The quantity of indicators for post-disaster recovery capacity; For the first Standardized values ​​for post-disaster recovery capacity indicators; For the first The weights corresponding to various disaster recovery capability indicators; Based on the repairability index of the disaster-bearing body corresponding to each historical disaster recovery sample, a damage rate mapping relationship is established:

[0048] in, The damage rate of the disaster-bearing body after it reaches the threshold of complete flooding; This is a mapping function established through statistical analysis or regression fitting.

[0049] The recovery time index characterizes the time required for a disaster-affected structure to recover from a damaged state to a normal usable state. It is determined based on the time interval between the disaster occurrence and the functional recovery time of the disaster-affected structure, and its expression is:

[0050] in, As a recovery time indicator; This refers to the actual recovery time. This is the preset maximum recovery time.

[0051] The recovery cost index is used to characterize the proportional relationship between the cost required to restore a disaster-affected area and the cost of complete reconstruction. Its expression is:

[0052] in, To restore cost indicators; This refers to the actual cost of post-disaster repair. Cost of complete reconstruction of the disaster-bearing body.

[0053] The functional recovery index is used to characterize the degree of post-disaster recovery of a disaster-affected body. It is determined based on the proportion of functions restored within a certain period after the disaster, and its expression is:

[0054] in, To restore cost indicators; For the restored functional value; Its pre-disaster functional value.

[0055] Resource security indicators are used to characterize the level of maintenance resources available during post-disaster recovery. They are determined based on the number of maintenance personnel, the supply capacity of maintenance materials, and the financial security capacity. Their expression is:

[0056] in, To restore cost indicators; To ensure the capabilities of maintenance personnel; For material supply capacity; To ensure financial security; Corresponding weights.

Claims

1. A flood risk assessment and damage calculation method based on tail risk state evolution, characterized in that, include: Collect time-stamped historical observation data of historical flood events in the target area; The historical observation data includes flood intensity indicators, loss data, and basic information on the affected areas. The flood intensity index of historical flood events is discretized using the flood intensity index classification threshold to obtain a flood risk dataset; Initialize the tail risk feature function using the flood risk dataset; Set up a sliding observation window. During the flood evolution, when the sliding observation window is updated, update the tail risk feature function based on the observation data of newly added flood events in the current sliding observation window. Based on the timescale relationship between the historical observation time length and the predicted time length of the future target time period, the tail risk characteristic function is transformed by timescale, and the frequency of occurrence of each flood intensity index level is predicted within the future target time period using the transformed tail risk characteristic function. Based on the exposure base, loss data and flood intensity index level of disaster-bearing bodies under various historical flood events, a nonlinear growth function is fitted to obtain the dynamic response relationship between flood intensity level and damage rate. The propagation process of floods under various flood intensity index levels is simulated using a flood evolution model to obtain the inundation range of the predicted flood scenario corresponding to each flood intensity index level. Based on the inundation range of each predicted flood scenario, the exposure base of the disaster-bearing bodies within the inundation range is summed to obtain the exposure base of the disaster-bearing bodies for each predicted flood scenario; The damage rate for each predicted flood scenario is determined by utilizing the dynamic response relationship between flood intensity level and damage rate. Multiply the frequency of occurrence, damage rate and exposure base of flood intensity index level corresponding to each predicted flood scenario by the product of the two factors to obtain the future damage scale corresponding to each predicted flood scenario. The total damage caused by the overall flood risk is obtained by summing the future damage scales corresponding to each predicted flood scenario.

2. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 1, characterized in that, The flood intensity index is the flood flow rate or flood velocity; the basic information data of the disaster-affected area includes the digital elevation model of the disaster-affected area, river geometric data, land use data, and building data.

3. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 1, characterized in that, The discretization specifically involves discretizing the flood intensity index of historical flood events into the lower limit value of the flood intensity index level of the historical flood events; the flood risk dataset... The expression is: in, For the first The lower limit of each flood intensity index level; To statistically analyze flood intensity indices based on discretized historical flood events. The frequency of occurrence.

4. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 1, characterized in that, The initialization of the tail risk feature function is specifically as follows: A set of distribution functions is constructed to fit the probability distribution law of flood intensity index, and the set of distribution functions includes several candidate probability distribution function models; Each candidate probability distribution function model was fitted using the flood risk tail distribution dataset; The fitting results of each candidate probability distribution function model are evaluated based on the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC), and the optimal probability distribution function model is selected. The optimal probability distribution function model obtained through screening is used to fit the tail distribution dataset of flood risk, and a tail risk feature function is constructed.

5. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 1, characterized in that, The tail risk feature function is updated based on the observation data of newly added flood events within the current sliding observation window, specifically as follows: Calculate the relative error between the observed data of newly added flood events within the sliding observation window and the current tail risk feature function prediction results: in, The relative error between newly observed flood event data and the prediction results of the tail risk characteristic function; When a new flood event occurs, the discrete value of the corresponding flood intensity index within the sliding observation window. The actual frequency of observations; Discrete values ​​of flood intensity index calculated based on the current tail risk characteristic function. The frequency of predictions; It is a set of discrete values ​​for flood intensity indices; The total number of elements in the discrete set of flood intensity index values; for The absolute value; Determine if the relative error is less than 10%. If yes, no action is taken; otherwise, introduce tail evolution weight coefficients to update the tail risk feature function. in, This is the updated tail risk feature function; The tail risk feature function before the update; Tail evolution weighting coefficients are used to characterize the change in the current flood sequence volatility relative to the historical overall volatility. This represents the standard deviation of the discrete values ​​of the flood intensity index for each flood event within the current sliding observation window; This represents the standard deviation of the discrete values ​​of flood intensity indices for each flood event within the total historical time span. This represents the discrete value of the flood intensity index.

6. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 1, characterized in that, The expression for the frequency of occurrence of each flood intensity index level within the target future time period is as follows: in, This represents the overall frequency of flood events within the target future timeframe. This is a dynamic time correction factor; This is the current tail risk characteristic function; This is the start time of the target time period; This is the end time of the target time period; This represents the number of dates that occurred within the current sliding observation window; This represents the number of dates within the timeframe of all historical observations that coincided with flood events. This represents the discrete value of the flood intensity index.

7. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 1, characterized in that, The dynamic response relationship between flood intensity level and damage rate is obtained as follows: The ratio of the exposure base and loss data of disaster-bearing bodies under each historical flood event is used as the physical damage rate under each historical flood event; based on the flood intensity index level of each historical flood event, the physical damage rate and the flood intensity index level are correlated to obtain the damage rate-intensity level response relationship dataset; By fitting a nonlinear growth function to the damage rate-intensity level response relationship dataset, the dynamic response relationship between flood intensity level and damage rate is obtained.

8. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 1, characterized in that, The expression for the dynamic response relationship between the flood intensity level and the damage rate is as follows: in, For the first Damage rate corresponding to each flood intensity level; , and All are fitted parameters; For the first Discrete values ​​corresponding to each flood intensity index level.

9. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 1, characterized in that, The determination of the damage rate for each predicted flood scenario is specifically as follows: when the flood intensity index corresponding to the predicted flood scenario does not reach the complete inundation threshold, the damage rate is determined based on the dynamic response relationship between the flood intensity level and the damage rate; when the flood intensity index corresponding to the predicted flood scenario reaches the preset flood intensity threshold corresponding to the complete inundation of the disaster-bearing body, the damage rate is determined based on the repairability of the disaster-bearing body.

10. The flood risk assessment and damage calculation method based on tail risk state evolution according to claim 9, characterized in that, The repairability of the disaster-bearing body is determined based on the post-disaster recovery capability index corresponding to the type of disaster-bearing body; the post-disaster recovery capability index includes at least one of the following: recovery time index, recovery cost index, functional recovery index, and resource guarantee index; the relationship between the repairability of the disaster-bearing body and the damage rate is established based on historical disaster recovery samples, including the following steps: Collect historical disaster recovery samples, which include indicators of the post-disaster recovery capacity of disaster-bearing bodies and historical damage rates under the highest upper limit of the flood intensity index level; The post-disaster recovery capacity indicators for each historical disaster recovery sample were standardized, and the repairability index of the disaster-bearing body corresponding to each historical disaster recovery sample was calculated separately: in, The index represents the repairability of the disaster-bearing body. The quantity of indicators for post-disaster recovery capacity; For the first Standardized values ​​for post-disaster recovery capacity indicators; For the first The weights corresponding to various disaster recovery capability indicators; Based on the repairability index of the disaster-bearing body corresponding to each historical disaster recovery sample, a damage rate mapping relationship is established: in, The damage rate of the disaster-bearing body after it reaches the threshold of complete flooding; This is a mapping function established through statistical analysis or regression fitting.

Images