Extreme rainfall risk prediction method, device, equipment, medium and product
Through multi-source data fusion and short-term trend forecasting, the recurrence period of extreme rainfall is calculated in real time, which solves the problem of lag or inaccurate traditional flood warning systems in the extreme rainfall process, and achieves timely early warning and dynamic adjustment of extreme rainfall risks.
Patent Information
- Application Number
- CN202510533806.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-25
- Publication Date
- 2025-08-01
AI Technical Summary
Traditional flood warning systems rely on rainfall data and fixed thresholds derived from long-term statistics, and cannot dynamically reflect the potential recurrence period level during extreme rainfall, resulting in late warning or inaccurate enough.
By obtaining multi-source rainfall data sequences, data fusion and short-term trend predictions, the average rainfall intensity of the target time window is calculated in real time, and the recurrence period is determined using interpolation method to achieve timely prediction of extreme rainfall risks.
It improves the timeliness and accuracy of extreme rainfall risk prediction, can dynamically adjust the warning level during rainfall, reduce early warning misjudgment, and improve the timeliness of flood control response.
Smart Images
Figure CN120409813A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of weather prediction, and particularly to a method, device, equipment, medium and product for predicting extreme rainfall risk. Background Art
[0002] With climate warming and frequent extreme weather, many regions are facing the flood risk brought by sudden and extreme rainfall. Traditional flood warning systems mostly rely on rainfall data obtained from long-term statistics and fixed thresholds for risk assessment. In the face of an ongoing extreme rainfall process, the warning is often lagged or inaccurate.
[0003] The Intensity Duration Frequency curve (IDF) is a commonly used tool in the fields of meteorology and flood control, which is used to represent the rainfall intensity corresponding to different durations and different return periods in a certain region. However, traditional methods mostly compare the measured rainfall with the IDF curve after the rainfall process ends, and cannot dynamically reflect the potential return period level during the rainfall, resulting in the inability to timely upgrade the flood warning level or take emergency measures. Summary of the Invention
[0004] The purpose of the present application is to provide a method, device, equipment, medium and product for predicting extreme rainfall risk, so as to dynamically predict the potential return period level during the rainfall and improve the timeliness of extreme rainfall risk prediction.
[0005] To achieve the above purpose, the present application provides the following solutions.
[0006] In the first aspect, the present application provides a method for predicting extreme rainfall risk, including:
[0007] Obtain a multi-source rainfall data sequence for the current time window;
[0008] Fuse the multi-source rainfall data at the same moment in the multi-source rainfall data sequence to obtain a fused rainfall sequence;
[0009] Based on the fused rainfall sequence and the actual rainfall data at each moment within the current time window, predict the rainfall data at each moment within the target time window, and form a predicted rainfall sequence for the target time window; the target time window is the next time window of the current time window;
[0010] Calculate the average rainfall intensity of the target time window according to the predicted rainfall sequence of the target time window;
[0011] Determine the return period of the current rainfall event based on the average rainfall intensity of the target time window.
[0012] Optionally, the formula for fusing multi-source rainfall data at the same moment in a multi-source rainfall data time series is:
[0013]
[0014] Among them, I fused (t) is the fused rainfall data at time t, I i (t) is the rainfall data at time t in the i-th data source, ω i is the weight of the i-th data source, and n is the number of data sources.
[0015] Optionally, the short-term trend prediction model is an exponential smoothing model;
[0016] The exponential smoothing model is:
[0017] in, is the predicted rainfall data at time t+Δt within the target time window, I fused (t) is the fused rainfall data at time t in the current time window, I t is the actual rainfall data at time t in the current time window, Δt is the length of the rolling time window, and α is the parameter of the exponential smoothing model.
[0018] Optionally, based on the average rainfall intensity in the target time window, the return period of the current rainfall event is determined, specifically including:
[0019] Based on the average rainfall intensity and target duration in the target time window, two return period curves adjacent to the current rainfall event are determined as a first return period curve and a second return period curve, respectively; the target duration is the total duration of the current rainfall event up to the last moment of the target time window;
[0020] Based on the first return period curve and the second return period curve, an interpolation method is used to determine the return period of the current rainfall event.
[0021] Alternatively, the formula for determining the return period of the current rainfall event using interpolation is:
[0022]
[0023] Among them, R est is the return period of the current rainfall event, I avg (t k ,t k +Δt) is the average rainfall intensity in the target time window, t k and t k +Δt are the starting and ending time of the target time window, Δt is the length of the rolling time window, I(D k, R1) is the average rainfall intensity I(D corresponding to the target duration D in the first return period curve k where I(D k , R2) is the average rainfall intensity I(D corresponding to the target duration D in the second return period curve, R2 is the return period of the second return period curve, and R1 is the return period of the first return period curve. k Optionally, the formula for calculating the average rainfall intensity of the target time window is:
[0024]
[0025]
[0026] where I avg (t k , t k +Δt) is the average rainfall intensity of the target time window, t k and t k +Δt are the start time and the end time of the target time window respectively, Δt is the length of the rolling time window, is the predicted rainfall data at the time t+Δt within the target time window.
[0027] In a second aspect, the present application provides an extreme rainfall risk prediction device. The extreme rainfall risk prediction device applies the above extreme rainfall risk prediction method. The extreme rainfall risk prediction device includes:
[0028] A multi-source rainfall data acquisition module for acquiring a multi-source rainfall data sequence of the current time window;
[0029] A multi-source rainfall data fusion module for fusing the multi-source rainfall data at the same moment in the multi-source rainfall data sequence to obtain a fused rainfall sequence;
[0030] A rainfall data prediction module for predicting the rainfall data at each moment within the target time window based on the fused rainfall sequence and the actual rainfall data at each moment within the current time window, and forming a predicted rainfall sequence of the target time window; the target time window is the next time window of the current time window;
[0031] An average rainfall intensity calculation module for calculating the average rainfall intensity of the target time window according to the predicted rainfall sequence of the target time window;
[0032] A return period determination module for determining the return period of the current rainfall event based on the average rainfall intensity of the target time window.
[0033] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the computer program to implement the above extreme rainfall risk prediction method.
[0034] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned extreme rainfall risk prediction method is implemented.
[0035] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the above-mentioned extreme rainfall risk prediction method is implemented.
[0036] According to the specific embodiments provided by the present application, the present application has the following technical effects.
[0037] The present application provides an extreme rainfall risk prediction method, device, equipment, medium and product. First, the present application obtains a multi-source rainfall data sequence of a current time window; fuses the multi-source rainfall data at the same moment in the multi-source rainfall data sequence to obtain a fused rainfall sequence; then, based on the fused rainfall sequence and the actual rainfall data at each moment within the current time window, predicts the rainfall data at each moment within a target time window to form a predicted rainfall sequence of the target time window; the target time window is the next time window of the current time window; calculates the average rainfall intensity of the target time window according to the predicted rainfall sequence of the target time window; determines the recurrence period of the current rainfall event based on the average rainfall intensity of the target time window. The present application can dynamically predict the rainfall data in the future time based on the fused rainfall data and the actual rainfall data during the rainfall process, and then determine the recurrence period based on the predicted rainfall data, improving the timeliness of extreme rainfall risk prediction. Description of the Drawings
[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0039] Figure 1 It is a schematic flowchart of an extreme rainfall risk prediction method provided by an embodiment of the present application.
[0040] Figure 2 It is a schematic structural diagram of a computer device provided by an embodiment of the present application. Detailed Embodiments
[0041] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.
[0042] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0043] Although there are some technologies that compare rainfall intensity with historical frequency, there is a lack of a method that can accurately and dynamically combine real-time rainfall data with historical IDF curves to predict the "once-in-a-year" or "recurrence period" level that the current rainfall may reach in a short period of time, and to link the flood warning system to respond immediately.
[0044] In an exemplary embodiment, a method for predicting extreme rainfall risk is provided, such as Figure 1 As shown, the process includes the following steps 101 to 105.
[0045] Step 101: Acquire a multi-source rainfall data sequence for the current time window.
[0046] Step 102 : Fusing the multi-source rainfall data at the same time in the multi-source rainfall data sequence to obtain a fused rainfall sequence.
[0047] Step 103: Based on the fused rainfall sequence and the actual rainfall data at each moment in the current time window, the rainfall data at each moment in the target time window is predicted to form a predicted rainfall sequence for the target time window; the target time window is the next time window of the current time window.
[0048] Step 104 : Calculate the average rainfall intensity of the target time window based on the predicted rainfall sequence of the target time window.
[0049] Step 105 : determining the recurrence period of the current rainfall event based on the average rainfall intensity in the target time window.
[0050] The implementation of the above steps 101 to 105 can achieve the following technical effects.
[0051] (1) Before the rainfall event ends, use real-time observation data and short-term forecasts to quickly and effectively estimate the possible recurrence period of the event.
[0052] (2) Realize the effective fusion and error correction of multi-source observation data (rain gauges, weather radars, numerical weather forecasts, etc.).
[0053] (3) Link the dynamic calculation results with the flood warning system to trigger or upgrade flood control response measures in real time.
[0054] Through technical means such as interpolation, multi-source data weighted fusion, and short-term trend prediction, the embodiment of this application constructs a real-time frequency assessment mechanism for extreme heavy rain, which has significant improvements in timeliness and accuracy, and can provide strong support for urban drainage and basin flood control.
[0055] In another exemplary embodiment, in step 101 above, rainfall data is obtained from multiple sources such as rain gauges and Numerical Weather Prediction (NWP). The rainfall data may include rainfall intensity and cumulative rainfall. In the embodiment of this application, the rainfall data refers to rainfall intensity.
[0056] In another exemplary embodiment, in step 102 above, multi-source data such as rain gauges and numerical weather forecasts are synthesized into a unified rainfall sequence I fused (t) in a weighted manner.
[0057] Multi-source weighted fusion can be achieved by the following formula:
[0058]
[0059] Where, I fused (t) is the fused rainfall data at time t, I i (t) is the rainfall data at time t in the i-th data source, ω i is the weight of the i-th data source, and n is the number of data sources.
[0060] In the embodiment of this application, ω i can be adaptively adjusted according to the credibility of different data sources under different rainfall intensity levels. The fused rainfall sequence is obtained for subsequent analysis.
[0061] In another exemplary embodiment, since the rainfall event usually has not ended, models such as exponential smoothing, ARIMA, or moving average can be used to predict the rainfall intensity in the next time window.
[0062] Exemplarily, the above short-term trend prediction model adopts the following exponential smoothing model:
[0063]
[0064] Where, is the predicted rainfall data at time t + Δt in the target time window, I fused (t) is the fused rainfall data at time t in the current time window, I tis the actual rainfall data at time t within the current time window, Δt is the length of the rolling time window, and α is the parameter of the exponential smoothing model.
[0065] In another exemplary embodiment, in step 104 above, according to the predicted rainfall sequence of the target time window, the average rainfall intensity of the target time window is calculated. The specific process is as follows:
[0066] (1) Set time windows of different lengths (short and long) to perform rolling cumulative transportation on the average rainfall intensity of each time window, and obtain the real-time updated average rainfall intensity, cumulative rainfall amount, and duration.
[0067] In the embodiments of the present application, multiple time window gears can be set (such as 10 minutes, 30 minutes, 1 hour, 3 hours, etc.), and the time window can be automatically switched to a shorter time window according to the current rainfall intensity to capture extreme instantaneous rainfall.
[0068] (2) After each time window ends, calculate the average rainfall intensity I avg (t k ,t k +Δt) and the cumulative rainfall amount P(t k ,t k +Δt), and update the target duration D k to the total duration of the current rainfall event up to the last moment of the target time window.
[0069]
[0070] Among them, I avg (t k ,t k +Δt) is the average rainfall intensity of the target time window, t k and t k +Δt are the start time and the last time of the target time window respectively, Δt is the length of the rolling time window, is the predicted rainfall data at time t+Δt within the target time window.
[0071] In another exemplary embodiment, after obtaining the average rainfall intensity, the IDF matching is performed using the average rainfall intensity to estimate the potential upgrade risk, which specifically includes the following steps 201-step 202.
[0072] Step 201, based on the average rainfall intensity and the target duration of the target time window, determine two recurrence interval curves adjacent to the current rainfall event, and use them as the first recurrence interval curve and the second recurrence interval curve respectively; the target duration is the total duration of the current rainfall event up to the last moment of the target time window.
[0073] Step 202: Based on the first return period curve and the second return period curve, determine the return period of the current rainfall event using an interpolation method.
[0074] The return period curve in step 201 is an empirical historical IDF curve, which is a known curve. The IDF parameters of this curve are obtained based on regression of rainfall observations over many years.
[0075]
[0076] Parameters such as C, d, m, and n are obtained by fitting historical rainfall data over many years. D represents duration, which represents the cumulative rainfall time. R represents the recurrence period (years). I represents the corresponding average rainfall intensity (mm / h).
[0077] In an exemplary embodiment, according to the observation value I avg (t k ,t k +Δt) and D k Determine whether the current rainfall event is between two typical return period curves (such as 10-year and 20-year return periods), and use linear or polynomial interpolation to estimate the return period of the current rainfall event.
[0078] The interpolation formula is as follows:
[0079]
[0080] Among them, R est is the return period of the current rainfall event, I avg (t k ,t k +Δt) is the average rainfall intensity in the target time window, t k and t k +Δt are the starting and ending time of the target time window, Δt is the length of the rolling time window, I(D k , R1) is the target duration D in the first return period curve k The corresponding average rainfall intensity, I(D k , R2) is the target duration D in the second return period curve k The corresponding average rainfall intensity, R2 is the return period of the second return period curve, and R1 is the return period of the first return period curve.
[0081] If the obtained R est If the set threshold is reached or exceeded (such as a once-in-50-year flood event), the flood warning will be automatically triggered or upgraded.
[0082] If linked with the flood evolution model, the recurrence period can be input into the flood evolution model to simulate possible flood peaks and inundation ranges, further improving the scientific nature of decision-making.
[0083] Innovation 1: Real-time dynamic assessment of ongoing rainfall events.
[0084] To address the problem that traditional IDF curve-based methods cannot instantly assess the return period during rainfall, resulting in delayed or inaccurate flood warnings, this paper uses a rolling time window to accumulate and analyze rainfall data in real time. Combined with short-term trend forecasting (exponential smoothing / moving average), this method repeatedly updates the range of possible return periods during rainfall, rapidly re-matching or interpolating the IDF curve to obtain an immediate estimate of rainfall frequency. This significantly improves the ability to identify ongoing rainstorms and the timeliness of warnings, buying critical time for flood prevention and emergency response deployment.
[0085] Innovation 2: Multi-source data fusion and error correction
[0086] A single rain gauge is affected by regional biases and equipment limitations, making it difficult to accurately reflect large-scale or localized severe convective rainfall conditions, which can easily lead to misjudgment of early warnings. This paper combines weighted fusion or bias correction of multi-source data such as automatic rain gauges and numerical weather forecasts, dynamically updates the rainfall intensity series, and reduces the impact of single data source inaccuracies on assessment results. This can more comprehensively and accurately reflect the actual distribution and intensity of rainfall areas, providing more reliable data input for subsequent IDF curve matching. This significantly reduces observation blind spots and error accumulation, improves the accuracy of capturing sudden local rainstorms, and makes the real-time estimated return period results more reliable.
[0087] Innovation 3: Automatic linkage flood warning and graded triggering
[0088] Traditional flood warning systems face the challenge of timely adjustment to warning levels, often facing escalating rainfall events. This paper links the real-time estimated rainfall return period to warning thresholds (e.g., 10-year, 20-year, or 50-year return periods). When a short-term rolling assessment indicates the return period has reached or exceeded a threshold, the flood warning level is automatically triggered or upgraded, and new rainfall intensity boundary conditions are pushed to the flood numerical model.
[0089] According to the specific embodiments provided in this application, this application has the following technical effects.
[0090] This application can realize real-time dynamic assessment of ongoing rainfall events.
[0091] To address the problem that traditional IDF curve-based methods cannot instantly assess the return period during rainfall, resulting in delayed or inaccurate flood warnings, this paper uses a rolling time window to accumulate and analyze rainfall data in real time. Combined with short-term trend forecasting (exponential smoothing / moving average), this method repeatedly updates the range of possible return periods during rainfall, rapidly re-matching or interpolating the IDF curve to obtain an immediate estimate of rainfall frequency. This significantly improves the ability to identify ongoing rainstorms and the timeliness of warnings, buying critical time for flood prevention and emergency response deployment.
[0092] This application can realize multi-source data fusion and error correction.
[0093] A single rain gauge is affected by regional biases and equipment limitations, making it difficult to accurately reflect large-scale or localized severe convective rainfall conditions, which can easily lead to misjudgment of early warnings. This paper combines weighted fusion or bias correction of multi-source data such as automatic rain gauges and numerical weather forecasts, dynamically updates the rainfall intensity series, and reduces the impact of single data source inaccuracies on assessment results. This can more comprehensively and accurately reflect the actual distribution and intensity of rainfall areas, providing more reliable data input for subsequent IDF curve matching. This significantly reduces observation blind spots and error accumulation, improves the accuracy of capturing sudden local rainstorms, and makes the real-time estimated return period results more reliable.
[0094] This application can realize automatic linkage flood warning and graded triggering.
[0095] Traditional flood warning systems face the challenge of timely adjustment to warning levels, often facing escalating rainfall events. This paper links the real-time estimated rainfall return period to warning thresholds (e.g., 10-year, 20-year, or 50-year return periods). When a short-term rolling assessment indicates the return period has reached or exceeded a threshold, the flood warning level is automatically triggered or upgraded, and new rainfall intensity boundary conditions are pushed to the flood numerical model.
[0096] Based on the same inventive concept, embodiments of the present application also provide an extreme rainfall risk prediction device for implementing the extreme rainfall risk prediction method described above. The solution provided by this device is similar to the solution described in the method described above. Therefore, the specific limitations of one or more embodiments of the extreme rainfall risk prediction device provided below can be found in the above-described limitations of the extreme rainfall risk prediction method and will not be further elaborated here.
[0097] In an exemplary embodiment, an extreme rainfall risk prediction device is provided, comprising:
[0098] Multi-source rainfall data acquisition module, used to obtain multi-source rainfall data sequence of the current time window;
[0099] A multi-source rainfall data fusion module, which is used to fuse multi-source rainfall data at the same moment in a multi-source rainfall data sequence to obtain a fused rainfall sequence;
[0100] A rainfall data prediction module, which is used to predict the rainfall data at each moment in a target time window based on the fused rainfall sequence and the actual rainfall data at each moment in the current time window, and form a predicted rainfall sequence for the target time window; the target time window is the next time window of the current time window;
[0101] An average rainfall intensity calculation module, which is used to calculate the average rainfall intensity of the target time window according to the predicted rainfall sequence of the target time window;
[0102] A recurrence period determination module, which is used to determine the recurrence period of the current rainfall event based on the average rainfall intensity of the target time window.
[0103] In an exemplary embodiment, a computer device is provided. The computer device can be a server or a terminal, and its internal structure diagram can be as Figure 2 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements an extreme rainfall risk prediction method.
[0104] Those skilled in the art can understand that Figure 2 the structure shown in
[0105] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program which, when executed by a processor, implements the steps in the above-described method embodiments.
[0106] In an exemplary embodiment, a computer program product is provided, including a computer program which, when executed by a processor, implements the steps in the above-described method embodiments.
[0107] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.
[0108] Those of ordinary skill in the art can understand that all or part of the processes in the above-described method embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it may include the processes of the above-described method embodiments. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application may include at least one of non-volatile and volatile memories. Non-volatile memories may include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAM), magnetoresistive random access memories (MRAM), ferroelectric random access memories (FRAM), phase change memories (PCM), graphene memories, etc. Volatile memories may include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.
[0109] In each of the embodiments provided in this application, the database involved may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., and is not limited thereto. In each of the embodiments provided in this application, the processor may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., and is not limited thereto.
[0110] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.
[0111] Specific examples are used in this article to elaborate on the principles and implementation manners of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to this application.
Claims
1. A method for predicting extreme rainfall risk, characterized in that, Including: Obtain the multi-source rainfall data sequence of the current time window; Fuse the multi-source rainfall data at the same moment in the multi-source rainfall data sequence to obtain a fused rainfall sequence; Based on the fused rainfall sequence and the actual rainfall data at each moment within the current time window, predict the rainfall data at each moment within the target time window, and form a predicted rainfall sequence for the target time window; the target time window is the next time window after the current time window; Calculate the average rainfall intensity of the target time window according to the predicted rainfall sequence of the target time window; Determine the recurrence interval of the current rainfall event based on the average rainfall intensity of the target time window.
2. The extreme rainfall risk prediction method according to claim 1, wherein The formula for fusing the multi-source rainfall data at the same moment in the multi-source rainfall data time series is: Among them, I fused (t) is the fused rainfall data at time t, and I i (t) is the rainfall data at time t in the i-th data source, ω i is the weight of the i-th data source, and n is the number of data sources.
3. The extreme rainfall risk prediction method according to claim 1, characterized in that The short-term trend prediction model is an exponential smoothing model; The exponential smoothing model is as follows: Among them, is the predicted rainfall data at time t + Δt within the target time window, I fused (t) is the fused rainfall data at time t within the current time window, I t is the actual rainfall data at time t within the current time window, Δt is the length of the rolling time window, and α is the parameter of the exponential smoothing model.
4. The extreme rainfall risk prediction method according to claim 1, wherein Determining the recurrence interval of the current rainfall event based on the average rainfall intensity of the target time window specifically includes: Based on the average rainfall intensity of the target time window and the target duration, determine two recurrence interval curves adjacent to the current rainfall event, and use them as the first recurrence interval curve and the second recurrence interval curve respectively; the target duration is the total duration of the current rainfall event up to the last moment of the target time window; Based on the first recurrence interval curve and the second recurrence interval curve, use the interpolation method to determine the recurrence interval of the current rainfall event.
5. The extreme rainfall risk prediction method according to claim 4, characterized in that The formula for using the interpolation method to determine the recurrence interval of the current rainfall event is: where, R est is the recurrence period of the current rainfall event, I avg (t k , t k +Δt) is the average rainfall intensity of the target time window, t k and t k +Δt are the start time and the last time of the target time window respectively, Δt is the rolling time window length, I(D k , R1) is the average rainfall intensity corresponding to the target duration D k in the first recurrence period curve, I(D k , R2) is the average rainfall intensity corresponding to the target duration D k in the second recurrence period curve, R2 is the recurrence period of the second recurrence period curve, and R1 is the recurrence period of the first recurrence period curve.
6. The extreme rainfall risk prediction method according to claim 1, wherein The formula for calculating the average rainfall intensity of the target time window is: Among them, I avg (t k , t k + Δt) is the average rainfall intensity of the target time window, t k and t k + Δt are the start time and the last time of the target time window respectively, and Δt is the rolling time window length, is the predicted rainfall data at the time of t + Δt within the target time window.
7. An extreme rainfall risk prediction device, characterized in that, The extreme rainfall risk prediction device applies the extreme rainfall risk prediction method described in any one of claims 1-6. The extreme rainfall risk prediction device includes: A multi-source rainfall data acquisition module for obtaining the multi-source rainfall data sequence of the current time window; A multi-source rainfall data fusion module for fusing the multi-source rainfall data at the same moment in the multi-source rainfall data sequence to obtain a fused rainfall sequence; A rainfall data prediction module for predicting the rainfall data at each moment within the target time window based on the fused rainfall sequence and the actual rainfall data at each moment within the current time window, and forming a predicted rainfall sequence for the target time window; the target time window is the next time window after the current time window; An average rainfall intensity calculation module for calculating the average rainfall intensity of the target time window according to the predicted rainfall sequence of the target time window; A recurrence interval determination module for determining the recurrence interval of the current rainfall event based on the average rainfall intensity of the target time window.
8. A computer device, comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that the processor executes the computer program to implement the extreme rainfall risk prediction method described in any one of claims 1-6.
9. 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 extreme rainfall risk prediction method described in any one of claims 1-6.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the extreme rainfall risk prediction method described in any one of claims 1-6.