Flood risk forecasting method and system considering index and weight uncertainty

Abstract

The invention provides a flood risk forecasting method and system considering index and weight uncertainty, and belongs to the technical field of risk analysis, and the method specifically comprises the steps: carrying out the determination of weight coefficients of flood monitoring indexes corresponding to different combinations according to the flood monitoring indexes of the historical flood process in different combinations, and carrying out the calculation of the weight coefficients of the flood monitoring indexes; and determining a stable monitoring index based on an analysis result of the weight coefficient, and determining a construction strategy of an influence model of the flood monitoring index based on stable monitoring index data and a change condition of the weight coefficient of the flood monitoring index between different combinations. According to the influence model and the rainfall data of the rainfall process before the current rainfall process, the flood risk prediction analysis method is determined, and the accuracy of flood risk prediction processing is improved.

Description

Technical Field

[0001] This invention belongs to the field of risk analysis technology, and in particular relates to a flood risk forecasting method and system that considers the uncertainty of indicators and weights. Background Technology

[0002] To achieve the analysis and processing of flood risk, the invention patent application CN202411713523.0, "A Dynamic Zoning Method and System for Flood Risk Based on Factor Multiple Optimization," constructs a risk assessment model and a weight optimization model for key flood risk assessment indicators. A hybrid genetic-local search algorithm using simulated annealing optimization is employed to calculate the optimal weight combination scheme, which is then input into the risk assessment model to obtain the dynamic zoning of flood risk in the study area. This provides a scientific and reasonable dynamic risk assessment. However, the following technical problems exist: As rainfall data changes, it inevitably leads to changes in indicators such as topography and soil type, which in turn causes changes in the weighting coefficients of different indicators that cause floods. Therefore, how to update and identify the weighting coefficients of different indicators to improve the accuracy of flood risk analysis and processing has become an urgent technical problem to be solved.

[0003] Therefore, there is an urgent need for a flood risk forecasting method and system that takes into account the uncertainty of indicators and weights. Summary of the Invention

[0004] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides a flood risk forecasting method that considers the uncertainty of indicators and weights, including: S1 uses historical flood data of a specific region as a basis to determine the changes in flood monitoring indicators between historical flood events. When the changes do not meet the requirements, the historical flood events are divided into multiple combinations based on the changes in the flood monitoring indicators. Based on the flood monitoring indicators of the historical flood events in different combinations, the weight coefficients of the flood monitoring indicators corresponding to different combinations are determined. Stable monitoring indicators are determined based on the analysis results of the weight coefficients. S2 determines the strategy for constructing an impact model of flood monitoring indicators based on the stable monitoring indicator data and the changes in the weight coefficients of flood monitoring indicators among different combinations. Based on the impact model and the rainfall data of the rainfall process before the current rainfall process, a flood risk prediction and analysis method is determined.

[0005] The beneficial effects of this invention are as follows: Based on stable monitoring indicator data, and considering the changes in the weight coefficients of flood monitoring indicators among different combinations, a strategy for constructing an impact model for flood monitoring indicators was determined. This approach identifies flood monitoring targets that require weight coefficient impact model construction from the perspective of the magnitude and risk of changes in the weight coefficients of flood monitoring targets, thereby improving the accuracy and reliability of the weight coefficients. Furthermore, by combining stable monitoring target data, a strategy for constructing differentiated impact models for flood monitoring indicators was also determined when there are differences in the constituent data of stable monitoring targets.

[0006] Based on the impact model and rainfall data from previous rainfall events, a predictive analysis method for flood risk is determined. This method not only considers the impact of rainfall data on the probability of changes in the weight coefficients of flood monitoring indicators, but also, by combining the impact model, assesses the reliability of the weight coefficients of the current flood monitoring targets. This, in turn, enables the determination of assessment schemes for the weight coefficients of other flood monitoring targets, thereby improving the accuracy of flood monitoring analysis results.

[0007] Furthermore, the historical flood data includes historical flood events in the specific region and the timing of different historical flood events.

[0008] Furthermore, the flood monitoring indicators include short-term heavy rainfall indicators, continuous cumulative rainfall indicators, high-sediment flow, water level critical value, flow matching degree, dam seepage flow, and soil moisture content.

[0009] Furthermore, the changes in flood monitoring indicators between historical flood events are determined based on the amount of change in flood monitoring indicators between different historical flood events, specifically based on the changes in monitoring data of flood monitoring indicators at the time of the flood.

[0010] Furthermore, determining that the aforementioned changes do not meet the requirements specifically includes: Based on the changes in flood monitoring indicators between the aforementioned historical flood events, the monitoring data of the flood monitoring targets at the time of the floods during the historical flood events are determined; Based on the monitoring data, determine the amount of change in the monitoring data of the historical flood event compared to the previous historical flood event; Based on the amount of change, determine whether the change meets the requirements.

[0011] Furthermore, the method for determining the flood risk prediction and analysis method is as follows: Based on the aforementioned impact model, flood monitoring indicators for constructing the impact model are determined; Based on the rainfall data, determine the rainfall data prior to the current rainfall event, and based on the rainfall data prior to the current rainfall event, determine the rainfall amount of the rainfall event prior to the current rainfall event; Based on the flood monitoring indicators processed by the aforementioned impact model construction and the rainfall amount of the previous rainfall process, a predictive analysis method for determining the flood risk is used.

[0012] In a second aspect, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the aforementioned flood risk forecasting method considering uncertainties in indicators and weights when running the computer program.

[0013] Other features and advantages will be set forth in the following description, and the objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0014] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0015] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0016] Figure 1 This is a flowchart of a flood risk forecasting method that considers the uncertainty of indicators and weights. Figure 2 This is a flowchart for determining whether changes do not meet the requirements; Figure 3 This is a flowchart illustrating the method for determining stability monitoring indicators; Figure 4 This is a flowchart illustrating the method for determining the construction strategy of the impact model of flood monitoring indicators; Figure 5 It is a framework diagram of a computer system. Detailed Implementation

[0017] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0018] This application identifies flood monitoring indicators that need to be included in the model by analyzing the changes in different flood monitoring indicators with rainfall data. This allows for targeted correction of the weight coefficients of the flood monitoring indicators, thereby improving the reliability of flood risk forecasting and avoiding the technical problem of the weight coefficients of flood monitoring indicators not being dynamically updated due to changes in topography caused by rainfall.

[0019] Example 1 like Figure 1 As shown, this application provides a flood risk forecasting method that considers the uncertainty of indicators and weights, specifically including: S1 uses historical flood data of a specific region as a basis to determine the changes in flood monitoring indicators between historical flood events. When the changes do not meet the requirements, the historical flood events are divided into multiple combinations based on the changes in the flood monitoring indicators. Based on the flood monitoring indicators of the historical flood events in different combinations, the weight coefficients of the flood monitoring indicators corresponding to different combinations are determined. Stable monitoring indicators are determined based on the analysis results of the weight coefficients. Furthermore, the historical flood data includes historical flood events in the specific region and the timing of different historical flood events.

[0020] Furthermore, the flood monitoring indicators include short-term heavy rainfall indicators, continuous cumulative rainfall indicators, high-sediment flow, water level critical value, flow matching degree, dam seepage flow, and soil moisture content.

[0021] Furthermore, the changes in flood monitoring indicators between historical flood events are determined based on the amount of change in flood monitoring indicators between different historical flood events, specifically based on the changes in monitoring data of flood monitoring indicators at the time of the flood.

[0022] Specifically, such as Figure 2 As shown, determining that the aforementioned changes do not meet the requirements specifically includes: Based on the changes in flood monitoring indicators between the aforementioned historical flood events, the monitoring data of the flood monitoring targets at the time of the floods during the historical flood events are determined; Based on the monitoring data, determine the amount of change in the monitoring data of the historical flood event compared to the previous historical flood event; Based on the amount of change, determine whether the change meets the requirements.

[0023] It is understood that, based on the amount of change, determining whether the change meets the requirements specifically includes: Based on the aforementioned amount of change, the average rate of change of the monitoring data of the flood monitoring target in different historical flood processes and the previous historical flood process is determined, and this average rate of change is used as the indicator rate of change. Based on the rate of change of flood monitoring indicators, determine whether the changes meet the requirements.

[0024] It should be noted that when there is a flood monitoring indicator whose rate of change is greater than the preset rate of change threshold, the change situation is determined to be unsatisfactory.

[0025] In one possible embodiment, the rate of change is determined based on the ratio of the absolute value of the change in the monitoring data of the flood monitoring target between the historical flood event and the previous historical flood event to the monitoring data of the historical flood event.

[0026] In another embodiment, determining that the change does not meet the requirements specifically includes: Based on the changes in flood monitoring indicators between the aforementioned historical flood events, the monitoring data of the flood monitoring targets at the time of the floods during the historical flood events are determined; Based on the monitoring data, determine the rate of change of the monitoring data of the historical flood event compared with that of the previous historical flood event; Based on the rate of change of the monitoring data of the historical flood event compared with the previous historical flood event, it is determined whether the change meets the requirements.

[0027] Specifically, when there are multiple historical flood events whose rate of change of monitoring data compared to the previous historical flood event does not meet the requirements, for example, when there are two or more historical flood events whose rate of change of monitoring data compared to the previous historical flood event is greater than 0.03, the changes in the flood monitoring indicators are relatively serious. Therefore, using a unified method for constructing weighting coefficients may not be able to accurately determine the weighting coefficients of the current flood monitoring indicators, and thus the changes are determined to not meet the requirements.

[0028] Furthermore, the historical flood events are divided into multiple combinations, specifically including: If the rate of change between the monitoring data of the flood monitoring target during the flood in the historical flood event and the monitoring data of the flood monitoring target during the previous historical flood event meets the requirements, then the historical flood event and the previous historical flood event are classified into the same group.

[0029] Furthermore, the weighting coefficients of the flood monitoring indicators are determined based on the monitoring data of historical flood processes in the combination, using correlation analysis methods to determine the weighting coefficients of different flood monitoring indicators.

[0030] It is understood that the correlation analysis methods mentioned include Pearson correlation analysis and Spearman rank correlation analysis.

[0031] Specifically, such as Figure 3 As shown, the method for determining the stability monitoring indicators is as follows: Based on the analysis results of the weighting coefficients, the weighting coefficients of flood monitoring indicators in different combinations are determined; Whether a flood monitoring indicator is a stable monitoring indicator is determined based on the similarity of the weighting coefficients of the flood monitoring indicators in different combinations.

[0032] It is understandable that when the weighting coefficients of flood monitoring indicators in different combinations are all within the same range, the water supply monitoring indicator is determined to be a stable monitoring indicator.

[0033] It should be noted that the value of the weighting coefficient is between 0 and 1. Specifically, based on a preset interval, the weighting coefficient is divided into multiple ranges at equal intervals, for example, the preset monitoring value is 0.05.

[0034] S2 determines the strategy for constructing an impact model of flood monitoring indicators based on the stable monitoring indicator data and the changes in the weight coefficients of flood monitoring indicators among different combinations. Based on the impact model and the rainfall data of the rainfall process before the current rainfall process, a flood risk prediction and analysis method is determined.

[0035] Specifically, such as Figure 4 As shown, the method for determining the construction strategy of the impact model of the flood monitoring indicators is as follows: Based on the stability monitoring index data, the composition of the stability monitoring index in the flood monitoring index is determined; Based on the variation of the weight coefficients of flood monitoring indicators among different combinations, the distribution data of the weight coefficients of the flood monitoring indicators in different ranges are obtained. Based on the distribution data of the weight coefficients of the flood monitoring indicators in different ranges and the composition of stable monitoring indicators in the flood monitoring indicators, the construction strategy of the impact model of the flood monitoring indicators is determined.

[0036] It is understandable that when the composition of stable monitoring indicators in the flood monitoring indicators does not meet the requirements, that is, when the number of stable monitoring indicators is small, if the influence model is not constructed for all monitoring indicators, the accuracy of the overall flood risk will inevitably be affected. Therefore, the influence model is constructed for all flood monitoring indicators.

[0037] Additionally, it can be understood that when the composition of the stable monitoring indicators in the flood monitoring indicators meets the requirements, the distribution data of the weight coefficients of the flood monitoring indicators in different ranges are determined. If the maximum value of its weight coefficient is within a preset range, that is, the weight coefficient of the flood monitoring indicator is small and its impact on flood risk is small, then it is determined that the flood monitoring indicators do not need to be processed by constructing an impact model.

[0038] Furthermore, if the maximum value of the weight coefficient of the flood monitoring index is not within the preset range, and if the number of weight coefficients of the flood monitoring index that exceed the preset number is greater than the preset number threshold, then it indicates that the distribution of the weight coefficients of the flood monitoring index is relatively dispersed. Therefore, it is determined that the flood monitoring index needs to undergo the construction of an impact model.

[0039] Additionally, it should be noted that if the number of flood monitoring indicators with a preset number of weight coefficients is not greater than a preset threshold, then the sum of the weight coefficients of the flood monitoring indicators after normalization processing of the monitoring data of different historical flood processes is used as the basis to determine the sum of the weight coefficients of the flood monitoring indicators that need to be processed for the construction of the influence model. If the sum of the weight coefficients of the flood monitoring indicators that need to be processed for the construction of the influence model is greater than the preset weight coefficient threshold in different historical flood processes, then the remaining flood monitoring indicators do not need to be processed for the construction of the influence model.

[0040] Furthermore, it is understood that if the sum of the weight coefficients of flood monitoring indicators requiring impact model construction is not greater than a preset weight coefficient threshold during different historical flood processes, the monitoring requirement value of the flood monitoring indicator is determined based on the weight coefficients of the flood monitoring indicators during historical flood processes where the sum of the weight coefficients of the flood monitoring indicators requiring impact model construction is not greater than the preset weight coefficient threshold. Specifically, if the monitoring requirement value of the flood monitoring indicator is greater than the preset requirement threshold, then the flood monitoring indicator is determined to require impact model construction; conversely, if the monitoring requirement value of the flood monitoring indicator is not greater than the preset requirement threshold, then the flood monitoring indicator is determined not to require impact model construction.

[0041] In one possible embodiment, the monitoring requirement value of the flood monitoring index is determined by the average value of the weight coefficients of the flood monitoring index during historical flood processes, where the sum of the weight coefficients of the flood monitoring indexes used in the influence model construction process is not greater than a preset weight coefficient threshold.

[0042] It should be noted that the impact model uses the rainfall amount between the previous historical flood event and the weight coefficients of the flood monitoring indicators of the previous historical flood event as inputs, and the weight coefficients of the flood monitoring indicators of the current historical flood event as outputs, thereby realizing the evaluation and processing of the impact of the weight coefficients corresponding to the rainfall data. In one possible embodiment, the impact model is constructed using one or more of BP neural networks and LSTM neural networks.

[0043] Example 2 Optionally, the method for determining the construction strategy of the impact model of the flood monitoring indicators is as follows: Based on the stability monitoring index data, the composition of the stability monitoring index in the flood monitoring index is determined; Based on the variation of the weight coefficients of flood monitoring indicators among different combinations, the distribution data of the weight coefficients of the flood monitoring indicators in different ranges are obtained, and the distribution discrete value of the flood monitoring indicators is determined based on the distribution data of the weight coefficients in different ranges. Based on the discrete distribution values ​​of the flood monitoring indicators and the composition of the stable monitoring indicators in the flood monitoring indicators, the strategy for constructing the impact model of the flood monitoring indicators is determined.

[0044] In one possible embodiment, the distributed discrete value is determined based on the number of ranges of weighted coefficients for the flood monitoring index. In another possible embodiment, it is determined based on the product of the number of ranges of weighted coefficients for the flood monitoring index and a preset scaling factor, for example, the preset scaling factor is 0.1.

[0045] It is understandable that when the composition of stable monitoring indicators in the flood monitoring indicators does not meet the requirements, that is, when the number of stable monitoring indicators is small, for example, when the proportion of stable monitoring indicators in all flood monitoring indicators is less than 0.5, if the influence model is not constructed for all monitoring indicators, the accuracy of the overall flood risk will inevitably be affected. Therefore, the influence model is constructed for all flood monitoring indicators.

[0046] It should also be noted that when the composition of the stable monitoring indicators in the flood monitoring indicators meets the requirements, if the distribution discrete value of the flood monitoring indicators is greater than the preset discrete threshold, for example, greater than 0.3, then it is determined that the flood monitoring indicators need to undergo the construction of an impact model.

[0047] Furthermore, if the distribution discrete value of the flood monitoring index is not greater than a preset discrete threshold, the model construction matching value of the flood monitoring index is determined based on the maximum value of the weight coefficient and the distribution discrete coefficient of the flood monitoring index. When the model construction matching value of the flood monitoring index is greater than the preset matching threshold, it is determined that the flood monitoring index needs to undergo influence model construction processing.

[0048] In one possible embodiment, the model construction matching value of the flood monitoring index is determined based on the maximum value of the weight coefficient of the flood monitoring index and the average value of the distribution dispersion coefficient. When the model construction matching value of the flood monitoring index is greater than 0.3, it is determined that the flood monitoring index needs to undergo influence model construction processing.

[0049] Additionally, it can be understood that when the model construction matching value of the flood monitoring index is not greater than the preset matching threshold, and when both the number of stable monitoring indicators requiring influence model construction processing and the number of flood monitoring indicators requiring influence model construction processing meet the requirements, then it is determined that the flood monitoring index does not need to undergo influence model construction processing.

[0050] In one possible embodiment, if the proportion of flood monitoring indicators requiring impact model construction to the total number of flood monitoring indicators and the proportion of stable monitoring indicators requiring impact model construction to the total number of stable monitoring indicators are both greater than 0.9, then it is determined that both the number of stable monitoring indicators requiring impact model construction and the number of flood monitoring indicators requiring impact model construction meet the requirements.

[0051] Furthermore, if either the number of stable monitoring indicators requiring impact model construction or the number of flood monitoring indicators requiring impact model construction does not meet the requirements, then all flood monitoring indicators that are not stable monitoring indicators will undergo impact model construction.

[0052] Optionally, the method for determining the flood risk prediction and analysis method is as follows: Based on the aforementioned impact model, flood monitoring indicators for constructing the impact model are determined; Based on the rainfall data, determine the rainfall data prior to the current rainfall event, and based on the rainfall data prior to the current rainfall event, determine the rainfall amount of the rainfall event prior to the current rainfall event; Based on the proportion of the flood monitoring indicators processed by the impact model construction in all flood monitoring targets, the model construction matching value is determined. Using the model construction matching value and the rainfall amount of the rainfall process before the current rainfall process, the flood risk prediction and analysis method is determined.

[0053] It is understandable that when the model matching value is greater than the preset matching threshold and all flood monitoring targets except for stable monitoring targets have constructed influence models, for example, when the value is 0.9, the risk weight coefficient of the flood monitoring target is determined by using the influence model. For stable monitoring targets that do not have an influence model, the risk weight coefficient of the flood monitoring target is determined by using its largest weight coefficient. The monitoring data obtained after normalizing the monitoring data of different flood monitoring targets are used to determine the flood risk prediction and analysis result by summing the risk weight coefficients of the monitoring data obtained after normalization.

[0054] Furthermore, when the model matching value is not greater than a preset matching threshold and all flood monitoring targets except for stable monitoring targets have not uniformly constructed influence models, the impact of flood monitoring indicators is determined based on the rainfall. Specifically, when the number of flood monitoring targets whose variation does not meet the requirements and whose influence models have not been constructed is greater than a preset target number threshold (e.g., greater than two), the risk weight coefficients of the flood monitoring targets are determined using influence models. For stable monitoring targets without influence models, the risk weight coefficients of the flood monitoring targets are determined using a preset method. The flood risk prediction analysis result is determined by summing the risk weight coefficients of the normalized monitoring data obtained from the normalized monitoring data.

[0055] It should be noted that, under the rainfall amount, the flood monitoring targets whose variation does not meet the requirements and for which an impact model has not been constructed are determined based on the variation of the weight coefficient of the flood monitoring targets among the combinations with the smallest absolute value of the deviation from the rainfall amount. Specifically, if the absolute value of the deviation between the rainfall amount and the rainfall amount within the interval between two adjacent combinations is the smallest, then the difference between the weight coefficient of the flood monitoring target and the previous combination is used to determine the variation of the weight coefficient of the flood monitoring target under the rainfall amount. When the variation is greater than 0.03, the flood monitoring target is determined to be a flood monitoring target whose variation does not meet the requirements and for which an impact model has not been constructed under the rainfall amount.

[0056] Furthermore, when the number of flood monitoring targets whose changes do not meet the requirements and for which an impact model has not been constructed is not greater than a preset target number threshold under the rainfall, the risk weight coefficient of the flood monitoring targets is determined using an impact model. For stable monitoring targets without an impact model, the risk weight coefficient of the flood monitoring targets is determined using a second preset method. The monitoring data obtained after normalizing the monitoring data of different flood monitoring targets are used as the sum of the risk weight coefficients of the normalized monitoring data to determine the flood risk prediction and analysis results.

[0057] Example 3 Secondly, such as Figure 5 As shown, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described flood risk forecasting method considering uncertainties in indicators and weights when running the computer program.

[0058] Furthermore, the method for determining the flood risk prediction and analysis method is as follows: Based on the aforementioned impact model, flood monitoring indicators for constructing the impact model are determined; Based on the rainfall data, determine the rainfall data prior to the current rainfall event, and based on the rainfall data prior to the current rainfall event, determine the rainfall amount of the rainfall event prior to the current rainfall event; Based on the flood monitoring indicators processed by the aforementioned impact model construction and the rainfall amount of the previous rainfall process, a predictive analysis method for determining the flood risk is used.

[0059] Specifically, when the flood monitoring indicators for which the impact model is constructed meet the requirements, that is, when all flood monitoring targets have had their impact models constructed, or when all flood monitoring targets except for stable monitoring targets have had their impact models constructed, the risk weight coefficients of the flood monitoring targets are determined using the impact models. For stable monitoring targets that do not have impact models, the risk weight coefficients of the flood monitoring targets are determined using their largest weight coefficients. The monitoring data obtained after normalizing the monitoring data of different flood monitoring targets are used as the sum of the risk weight coefficients of the normalized monitoring data to determine the flood risk prediction and analysis results.

[0060] Furthermore, when the flood monitoring indicators processed for the construction of the influence model do not meet the requirements, the number of flood monitoring indicators that have not undergone the construction of the influence model is determined. If the number of flood monitoring targets that have not undergone the construction of the influence model is greater than the preset threshold for the number of monitoring targets, the risk weight coefficient of the flood monitoring targets is determined using the influence model. If there are no stable monitoring targets with influence models, the risk weight coefficient of the flood monitoring targets is determined using a preset method. The monitoring data obtained after normalizing the monitoring data of different flood monitoring targets are used as the sum of the risk weight coefficients of the monitoring data obtained after normalization to determine the prediction and analysis results of the flood risk.

[0061] Additionally, it can be understood that if the number of flood monitoring targets that have not undergone impact model construction exceeds a preset threshold for the number of monitoring targets, the weights of the flood monitoring targets that have not undergone impact model construction are determined based on their types. If the weights of the flood monitoring targets that have not undergone impact model construction do not meet the requirements, the risk weight coefficients of the flood monitoring targets are determined using the impact model. For stable monitoring targets that do not have an impact model, the risk weight coefficients of the flood monitoring targets are determined using a preset method. The monitoring data obtained after normalizing the monitoring data of different flood monitoring targets are used to determine the flood risk prediction and analysis results by summing the risk weight coefficients of the normalized monitoring data.

[0062] It should be noted that the weight of the flood monitoring target that has not undergone the construction of the impact model is determined by the sum of the weight values ​​of the flood monitoring target that has not undergone the construction of the impact model. The weight values ​​of the flood monitoring target that has not undergone the construction of the impact model are determined according to the type of the flood monitoring target that has not undergone the construction of the impact model. The weight value of the stable monitoring target is less than that of the flood monitoring target that is not a stable monitoring target. In one possible embodiment, the weight values ​​are 0.1 and 0.2, respectively.

[0063] Additionally, it can be understood that when the weights of flood monitoring targets that have not undergone impact model construction meet the requirements, the rainfall amount of the rainfall process preceding the current rainfall process, i.e., the rainfall amount of the rainfall process between the most recent historical flood process, is determined. Based on the rainfall amount, the impact of flood monitoring indicators is determined. Where the number of flood monitoring targets that have not undergone impact model construction and whose changes do not meet the requirements under the rainfall amount is greater than a preset target number threshold, the risk weight coefficient of the flood monitoring targets is determined using the impact model. For stable monitoring targets that do not have an impact model, the risk weight coefficient of the flood monitoring targets is determined using a preset method. The monitoring data obtained after normalizing the monitoring data of different flood monitoring targets is used as the sum of the risk weight coefficients of the normalized monitoring data to determine the prediction and analysis results of the flood risk.

[0064] Furthermore, when the number of flood monitoring targets whose changes do not meet the requirements and for which an impact model has not been constructed is not greater than a preset target number threshold under the rainfall, the risk weight coefficient of the flood monitoring targets is determined using an impact model. For stable monitoring targets without an impact model, the risk weight coefficient of the flood monitoring targets is determined using a second preset method. The monitoring data obtained after normalizing the monitoring data of different flood monitoring targets are used as the sum of the risk weight coefficients of the normalized monitoring data to determine the flood risk prediction and analysis results.

[0065] It should be noted that the first preset method is based on the change in the weight coefficient of the flood monitoring target under the rainfall, combined with the maximum value of the weight coefficient of the flood monitoring indicators in different combinations, and the risk weight coefficient of the flood monitoring target is determined based on the sum of the change and the maximum value of the weight coefficient. The second preset method is based on the change in the weight coefficient of the flood monitoring target under the rainfall, combined with the weight coefficient of the flood monitoring indicators in the combination corresponding to the most recent historical flood process, and the risk weight coefficient of the flood monitoring target is determined based on the sum of the change and the weight coefficient of the monitoring data.

[0066] Furthermore, the change in the weight coefficient of the flood monitoring target under the rainfall is determined based on the change in the weight coefficient of the flood monitoring target among the combinations with the smallest absolute value of the deviation from the rainfall. If the absolute value of the deviation from the rainfall within the interval between two adjacent combinations is the smallest, then the difference in the weight coefficient of the flood monitoring target with respect to the previous combination is used to determine the change in the weight coefficient of the flood monitoring target under the rainfall.

[0067] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0068] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0069] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A flood risk forecasting method considering the uncertainty of indicators and weights, characterized in that, Specifically, it includes: Based on historical flood data of a specific region, the changes in flood monitoring indicators between historical flood events are determined. When the changes do not meet the requirements, the historical flood events are divided into multiple combinations according to the changes in the flood monitoring indicators. Based on the flood monitoring indicators of the historical flood events in different combinations, the weight coefficients of the flood monitoring indicators corresponding to different combinations are determined. Stable monitoring indicators are determined based on the analysis results of the weight coefficients. Based on the stable monitoring index data, and based on the changes in the weight coefficients of flood monitoring indicators among different combinations, the strategy for constructing the impact model of flood monitoring indicators is determined. Based on the aforementioned impact model and rainfall data from previous rainfall events, a predictive analysis method for flood risk is determined.

2. The flood risk forecasting method considering the uncertainty of indicators and weights as described in claim 1, characterized in that, The historical flood data refers to the historical flood events in the specific region and the times of different historical flood events.

3. The flood risk forecasting method considering the uncertainty of indicators and weights as described in claim 1, characterized in that, The flood monitoring indicators include short-term heavy rainfall indicators, continuous cumulative rainfall indicators, high-sediment flow, water level critical value, flow matching degree, dam seepage flow, and soil moisture content.

4. The flood risk forecasting method considering the uncertainty of indicators and weights as described in claim 1, characterized in that, The changes in flood monitoring indicators between historical flood events are determined based on the amount of change in flood monitoring indicators between different historical flood events.

5. The flood risk forecasting method considering the uncertainty of indicators and weights as described in claim 1, characterized in that, The determination that the aforementioned changes do not meet the requirements specifically includes: Based on the changes in flood monitoring indicators between the aforementioned historical flood events, the monitoring data of the flood monitoring targets at the time of the floods during the historical flood events are determined; Based on the monitoring data, determine the amount of change in the monitoring data of the historical flood event compared to the previous historical flood event; Based on the amount of change, determine whether the change meets the requirements.

6. The flood risk forecasting method considering the uncertainty of indicators and weights as described in claim 5, characterized in that, Based on the amount of change, determining whether the change meets the requirements specifically includes: Based on the aforementioned amount of change, the average rate of change of the monitoring data of the flood monitoring target in different historical flood processes and the previous historical flood process is determined, and this average rate of change is used as the indicator rate of change. Based on the rate of change of flood monitoring indicators, determine whether the changes meet the requirements.

7. The flood risk forecasting method considering the uncertainty of indicators and weights as described in claim 1, characterized in that, The historical flood events are divided into multiple combinations, specifically including: If the rate of change between the monitoring data of the flood monitoring target during the flood in the historical flood event and the monitoring data of the flood monitoring target during the previous historical flood event meets the requirements, then the historical flood event and the previous historical flood event are classified into the same group.

8. The flood risk forecasting method considering the uncertainty of indicators and weights as described in claim 1, characterized in that, The weighting coefficients of the flood monitoring indicators are determined based on the monitoring data of historical flood processes in the combination, using correlation analysis methods to determine the weighting coefficients of different flood monitoring indicators.

9. The flood risk forecasting method considering the uncertainty of indicators and weights as described in claim 1, characterized in that, The method for determining the flood risk prediction and analysis method is as follows: Based on the aforementioned impact model, flood monitoring indicators for constructing the impact model are determined; Based on the rainfall data, determine the rainfall data prior to the current rainfall event, and based on the rainfall data prior to the current rainfall event, determine the rainfall amount of the rainfall event prior to the current rainfall event; Based on the flood monitoring indicators processed by the aforementioned impact model construction and the rainfall amount of the previous rainfall process, a predictive analysis method for determining the flood risk is used.

10. A computer system, comprising: A memory and processor connected by communication, and a computer program stored in the memory and capable of running on the processor, characterized in that, when the processor runs the computer program, it executes a flood risk forecasting method considering the uncertainty of indicators and weights as described in any one of claims 1-9.

Images