Accumulated water point critical rainfall dynamic assessment and accumulated water risk prediction method

Abstract

The invention discloses a ponding point critical rainfall dynamic assessment and ponding risk prediction method, relates to the technical field of critical rainfall dynamic assessment and ponding risk prediction, and solves the technical problems that a traditional mechanism model is limited, when sample data is less and the ponding point state changes, timely identification cannot be performed, and the ponding prediction accuracy is low. According to the method, each rainfall field is dynamically tracked, accumulated water information statistical data of the accumulated water points are obtained in an accumulated mode and serve as sample data, rainfall time period statistical data of the accumulated water points without accumulated water are obtained and serve as the basis for recognizing the state change of the accumulated water points, a four-level critical rainfall threshold value of the accumulated water points is set, and three critical rainfall indexes are selected; extracting critical rainfall threshold values of each level and each index through a dynamic maximum and minimum value taking method; according to the method, the states of the ponding points can be identified and distinguished more meticulously, the ponding critical rainfall threshold is dynamically obtained, and the accuracy and the real-time performance of urban inland inundation ponding prediction can be improved.

Description

Technical Field

[0001] The invention belongs to the field of urban hydrology, and relates to a critical rainfall dynamic assessment and waterlogging risk prediction technology, and specifically is a critical rainfall dynamic assessment and waterlogging risk prediction method for a waterlogging point. Background Art

[0002] Urban waterlogging refers to the phenomenon of waterlogging disasters in cities due to heavy or continuous rainfall exceeding the city's drainage capacity. The main causes of urban waterlogging are: heavy or continuous rainfall; insufficient urban drainage capacity caused by imperfect urban drainage facilities, aging pipe networks and low emission standards; a large number of hard pavements in cities, such as asphalt roads and cement roads, have poor water permeability during rainfall, are not easy to infiltrate, and are prone to forming road surface waterlogging. Urban waterlogging will have a significant impact on people's production and life, traffic and economic development, and may even cause significant loss of life and property. Establishing a real-time prediction and early warning method and system for waterlogging is of great significance for early warning of waterlogging and improving waterlogging prevention and control capabilities.

[0003] Real-time prediction and warning of urban waterlogging is to predict and issue warnings for urban waterlogging scenarios before heavy rainfall or during continuous rainfall. Currently, the following three technical means are more advanced in the industry:

[0004] 1. Real-time prediction technology for urban waterlogging based on urban hydrological and hydrodynamic models. This technology is based on the physical property data of urban drainage pipes, terrain, underlying surface, and engineering facilities. It builds a refined drainage process mechanism model that covers urban surface runoff, pipe networks, rivers, surface waterlogging, and hydraulic structures. By accessing rainfall forecast data, real-time data, and dynamic monitoring data of drainage facilities, it can achieve refined, real-time rolling predictions of urban waterlogging. This technology is highly dependent on the quality and integrity of basic data, defect data, dynamic monitoring data, and historical sample data, but it is difficult to feed back defect data such as pipe network siltation, disconnection, and damage to the model in real time, making it difficult for the model to reflect all actual conditions on site.

[0005] 2. Real-time prediction technology for waterlogging based on artificial intelligence algorithms. This technology uses various types of neural networks and machine learning algorithms as basic algorithms. It uses the time series data of multiple historical rainstorms and waterlogging as sample data to train the algorithm, directly establishes a nonlinear relationship between rainfall and waterlogging depth, and realizes real-time prediction of waterlogging by accessing rainfall forecast data and actual data. This technology has a fast calculation speed, but does not consider the process mechanism of rainfall-waterlogging and the state changes of waterlogging points, and has high requirements for the amount of historical sample data.

[0006] 3. Real-time prediction technology for urban waterlogging by coupling urban hydrological and hydrodynamic models with artificial intelligence algorithms. This technology first establishes a set of urban hydrological and hydrodynamic models that meet the accuracy requirements, and calculates a large amount of rainfall-waterlogging case data through the model, which is used as training samples for the artificial intelligence algorithm. The algorithm accesses rainfall forecast data and actual data to achieve real-time prediction of urban waterlogging. This technology has a fast calculation speed and can obtain a large amount of sample data, but it relies too much on the results of the hydrological and hydrodynamic model and has the problem of error accumulation.

[0007] At present, most real-time and rapid prediction methods for urban waterlogging are unable to meet the requirements of timely identifying waterlogging conditions and accurately predicting waterlogging risks when there is little sample data or when the status of waterlogging points changes, which increases the risk of waterlogging disasters in cities.

[0008] Therefore, the present invention discloses a method for dynamically evaluating the critical rainfall at a waterlogging point and predicting waterlogging risk, which is used to solve the above technical problems. Summary of the invention

[0009] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a method for dynamic evaluation of critical rainfall at waterlogging points and prediction of waterlogging risk, which is used to solve the technical problems that due to the limitations of traditional mechanism models, when there is a small amount of sample data and the waterlogging point state changes cannot be identified in time, and the accuracy of waterlogging prediction is low. The present invention dynamically tracks each rainfall, accumulates statistical data of waterlogging information at waterlogging points, and uses this as sample data to obtain statistical data of rainfall periods when there is no waterlogging at the waterlogging points. As a basis for identifying changes in the state of waterlogging points, four-level critical rainfall thresholds for waterlogging points are set, and three critical rainfall indicators are selected. The critical rainfall thresholds of each level and each indicator are extracted by dynamic maximum value method to solve the above problems.

[0010] To achieve the above-mentioned object, the first aspect of the present invention provides a method for dynamically evaluating critical rainfall at a waterlogging point and predicting waterlogging risk, comprising:

[0011] Obtain the rainfall start time and rainfall end time of each rainfall, calculate the rainfall index for each rainfall, and store the rainfall start time, rainfall end time and rainfall index of each rainfall in the total database;

[0012] Establish a waterlogging point database, calculate and dynamically update the critical rainfall threshold based on the waterlogging point database to obtain a four-level threshold table of critical rainfall at the waterlogging point; wherein the waterlogging point database includes a waterlogging point waterlogging information sample database and a waterlogging point non-waterlogging rainfall period database;

[0013] The time and risk of waterlogging are predicted based on the four-level threshold table of critical rainfall at waterlogging points.

[0014] Preferably, the obtaining of the rainfall start time and rainfall end time of each rainfall includes:

[0015] A1: Determine whether the water depth monitoring equipment at the water accumulation point has an alarm; if yes, record the time t0 when the monitoring alarm is generated and jump to A2; if no, do nothing;

[0016] A2: The nth hour before the monitoring alarm generation time t0 is recorded as t 0-n , traverse by time t 0-n The rainfall data from time to time t0 is used to determine whether a rainfall greater than V can be found in this traversal process. min rainfall data; yes, the first one greater than V min The collection time of rainfall data is marked as the rainfall start time t s , and jump to A3; No, stop analyzing the rainfall start time and rainfall end time corresponding to this alarm; where the value of n is determined according to the area size of the drainage unit, V min Indicates the minimum resolution for rainfall monitoring;

[0017] A3: t s The time is the first starting time, and the rainfall data from the latest first starting time to n hours before the first starting time are traversed again. When a rainfall greater than V can be found in this traversal process min When the rainfall data is available, use the first value greater than V min The collection time of rainfall data is relative to the rainfall start time t s Make updates;

[0018] A4: Repeat step A3 until no rainfall greater than V is found in the rainfall data from the latest first starting point to n hours before the first starting point. min When the rainfall data is output, the rainfall start time t s ;

[0019] A5: The time after the monitoring alarm generation time t0 is recorded as t 0+1 , traverse from time t0 to t 0+1 The rainfall data at each moment is used to determine whether a rainfall greater than V can be found during this traversal. min rainfall data; yes, the last one greater than V min The collection time of rainfall data is marked as the end time of rainfall t e , with t e The time is the second starting time; otherwise, t 0+1 The time is the second starting time;

[0020] A6: traverse the rainfall data from the latest second starting point to 1 hour after the second starting point again; when a rainfall greater than V can be found in this traversal processmin When the rainfall data is greater than V, use the last min The collection time of rainfall data is relative to the rainfall end time t e Update and use the latest rainfall end time t e Update the second starting point time; when no value greater than V is found in this traversal process min When the rainfall data is obtained, the second starting point time is updated with the time 1 hour after the second starting point time used in this traversal;

[0021] A7: Repeat step A6 until the latest rainfall ends at time t e As the starting point, no value greater than V can be found in the next n hours. min When the rainfall data is obtained; determine the rainfall end time t e Is it empty? If yes, the monitoring alarm generation time t0 is used as the rainfall end time t e Output; No, output the latest rainfall end time t e .

[0022] Preferably, the calculation of the rainfall index for each rainfall includes:

[0023] B1: Extract the rainfall start time t of this rainfall s and the rainfall end time t e , judge t e -t s Is it greater than 3 hours? If yes, calculate the rainfall every 3 hours using the sliding time window and compare to get t s Time t e Maximum 3-hour rainfall between and 3h ; No, jump to B2;

[0024] B2: Determine t e -t s Is it greater than 2 hours and 30 minutes? If yes, use t s Time t e The accumulated rainfall between and is taken as the maximum 3-hour rainfall I 3h ; No, set the maximum 3-hour rainfall I 3h Set to null value;

[0025] B3: Extract the rainfall start time t of this rainfall s and the rainfall end time t e , judge t e -t s Is it greater than 1 hour? If yes, calculate the rainfall every 1 hour using the sliding time window and compare to get t s Time t eMaximum hourly rainfall between and h ; No, jump to B4;

[0026] B4: Determine t e -t s Is it greater than 40 minutes? Yes, in t s Time t e The accumulated rainfall between and is taken as the maximum hourly rainfall I h ; No, set the maximum hourly rainfall I h Set to null value.

[0027] Preferably, the establishment of a water accumulation point database includes:

[0028] Extract all rainfall data stored in the database corresponding to the current waterlogging point, and extract the rainfall period t of each rainfall data i s ~t i e , extract each rainfall period t i s ~t i e The maximum water depth h of the corresponding water point i max ; where i is the recording order of rainfall data, rainfall period t i s ~t i e The time range is from time t i s At time t i e ;

[0029] The maximum water depth h i max The rainfall period of not less than 0.05 meters i s ~t i e Mark as target time period, extract the maximum water depth h in each target time period i max The waterlogging level is located, and the critical rainfall time and critical rainfall duration of the waterlogging level are obtained by interpolation method; among them, the waterlogging level includes level IV waterlogging level: 0<h max ≤h 4,max , Grade III water accumulation level: h 4,max <h max ≤h 3,max , Level II water accumulation level: h 3,max <h max ≤h 2,max , Grade I waterlogging level: h 2,max <hmax ;

[0030] The maximum water depth h in the target time period corresponding to each waterlogging point is obtained. i max , water accumulation level, rainfall start time t i s , Rainfall end time t i e , critical rainfall time, critical rainfall duration, and the maximum hourly rainfall obtained according to the critical rainfall duration I h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i The water accumulation information is recorded in a water accumulation point water accumulation information statistical table to form a water accumulation point water accumulation information sample database;

[0031] The maximum water depth h i max The rainfall period is less than 0.05 meters. i s ~t i e Mark as non-target time period, and calculate the maximum hourly rainfall I in each non-target time period according to steps A1 to A7 and steps B1 to B4. h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i ＜0.05 ;

[0032] The maximum hourly rainfall I h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i ＜0.05 The data are recorded in the statistical table of rainfall periods without water accumulation at water accumulation points, forming a database of rainfall periods without water accumulation at water accumulation points.

[0033] Preferably, the method of obtaining the critical rainfall moment and critical rainfall duration of the water accumulation level by using the interpolation method includes:

[0034] Extract the waterlogging level and determine whether it is level IV. If yes, use the interpolation method to obtain the maximum waterlogging depth h in this record. i max The first time when the rainfall is equal to 0.05 meters is marked as the critical rainfall moment of Grade IV flooding level t i 0.05 ; No, use the interpolation method to obtain the moment when the water depth of the waterlogging point reaches the minimum value of the waterlogging level, and mark the moment as the critical rainfall moment of the waterlogging level; among them, the critical rainfall moment of level III waterlogging level is marked as t i (h4,max-Δh3,cr), the critical rainfall moment of level II flooding is marked as t i (h3,max-Δh2,cr) , the critical rainfall moment of level I flooding is marked as t i (h2,max-Δh1,cr) ;

[0035] Get the time interval of the critical rainfall moment [t j-1 ,t j ], and the time interval [t j-1 ,t j ] Total rainfall V j , calculate t using interpolation j-1 The rainfall Vz in the time period between the critical rainfall moment and the critical rainfall moment; where the time interval [t j-1 ,t j ] is the collection time of rainfall data, and z in rainfall Vz is the classification number of waterlogging level;

[0036] Calculate the rainfall start time t i s Critical rainfall time to level IV flooding level t i 0.05 The critical rainfall duration Δt i 4. Calculate the rainfall start time t i s Critical rainfall time to level III flooding level t i (h4,max-Δh3,cr) The critical rainfall duration Δt i 3. Calculate the rainfall start time t i s Critical rainfall time to level II flooding level t i (h3,max-Δh2,cr) The critical rainfall duration Δt i 2. Calculate the rainfall start time t i s The critical rainfall moment for level I flooding is marked as t i (h2,max-Δh1,cr) The critical rainfall duration Δt i 1.

[0037] Preferably, the calculation and dynamic updating of the critical rainfall threshold to obtain a four-level threshold table of critical rainfall at waterlogging points includes:

[0038] According to the latest waterlogging information sample database, the rainfall periods with the same waterlogging level are compared and the minimum maximum hourly rainfall I min h,z , minimum maximum 3-hour rainfall I min 3h,z and the minimum accumulated rainfall V min zAs the initial value of the critical rainfall threshold of the corresponding level, and record the rainfall period (t i s ~t i e ) min w,z , the minimum maximum hourly rainfall I in each waterlogging level at the waterlogging point min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min And the rainfall period corresponding to each minimum value (t i s ~t i e ) min w,z Recorded in the four-level threshold table of critical rainfall at waterlogging points; among them, rainfall period (t i s ~t i e ) min w,z Where w is the classification number of rainfall, and z is the classification number of waterlogging level;

[0039] The critical rainfall thresholds are dynamically updated to obtain a four-level threshold table of critical rainfall at waterlogging points.

[0040] Preferably, the dynamic updating of the critical rainfall threshold to obtain a four-level threshold table of critical rainfall at waterlogging points includes:

[0041] C1: Determine whether the IV-level threshold stored in the IV-level threshold table of critical rainfall at waterlogging points is not empty; if yes, jump to C2; if no, jump to C5; among which, non-empty means there is a specific value;

[0042] C2: Extract the rainfall period corresponding to the stored IV level threshold, mark all rainfall periods after the rainfall period corresponding to the stored IV level threshold as reference rainfall periods in the latest rainfall period database without water accumulation, and count the maximum value I of the maximum hourly rainfall in the reference rainfall period. h lat_max and the second largest value I h lat_semax , the maximum value of the maximum rainfall in 3 hours I 3h lat_max and the second largest value I 3h lat_semax , the maximum value of the accumulated rainfall V lat_max and the next largest value V lat_semax ;

[0043] C3: If the maximum value and the second maximum value are ≥ K times the existing IV level threshold of the corresponding indicator, that is:

[0044] I h lat_max ,I h lat_semax ≥K×I min h,Ⅳ ;

[0045] or I 3h lat_max ,I 3h lat_semax ≥K×I min 3h,Ⅳ ;

[0046] or V lat_max 、V lat_semax ≥K×V min Ⅳ ;

[0047] Among them, K is a proportional adjustment coefficient greater than 1;

[0048] C4: Determine whether the maximum value and the second maximum value in the three formulas of C3 are ≥ K times the existing IV-level threshold of the corresponding indicator. If yes, mark the earliest rainfall period in the rainfall period corresponding to the established formula as the earliest rainfall period. In the waterlogging point waterlogging information sample database, if the minimum maximum hourly rainfall I with waterlogging level of IV after the earliest rainfall period can be found, min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min , the corresponding minimum value is used as the latest IV level threshold; if the minimum maximum hourly rainfall I with water accumulation level of IV after the earliest rainfall period cannot be found, min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min , then directly use the maximum value of the maximum hourly rainfall in the reference rainfall period I h lat_max , the maximum value of the maximum rainfall in 3 hours I 3h lat_max , the maximum value of the accumulated rainfall V lat_max As the latest level IV threshold;

[0049] No, do not update the existing Level IV threshold;

[0050] C5: Query the latest database of rainfall periods with or without water accumulation, and extract the maximum hourly rainfall I maxh,<0.05 , Maximum 3-hour rainfall I max 3h,<0.05 and the maximum accumulated rainfall V max <0.05 , and the rainfall period corresponding to each maximum value (t i s ~t i e ) max w,<0.05 ; Determine whether the maximum value is not less than the minimum value of the threshold corresponding to each level, and the rainfall period corresponding to the maximum value is greater than the rainfall period corresponding to the minimum value of the threshold, that is:

[0051] I max h,<0.05 ≥min(I min h,Ⅲ ,I min h,Ⅱ ,I min h,Ⅰ ), and (t i s ~t i e ) max h,<0.05 >min((t i s ~t i e ) min h,Ⅲ ,(t i s ~t i e ) min h,Ⅱ ,(t i s ~t i e ) min h,Ⅰ );

[0052] or I max 3h,<0.05 ≥min(I min 3h,Ⅲ ,I min 3h,Ⅱ ,I min 3h,Ⅰ ), and (t i s ~t i e ) max 3h,<0.05 >min((t i s ~t ie ) min 3h,Ⅲ ,(t i s ~t i e ) min 3h,Ⅱ ,(t i s ~t i e ) min 3h,Ⅰ );

[0053] or V max <0.05 ≥min(V min Ⅲ ,V min Ⅱ ,V min Ⅰ ), and (t i s ~t i e ) max V,<0.05 >min((t i s ~t i e ) min V,Ⅲ ,(t i s ~t i e ) min V,Ⅱ ,(t i s ~t i e ) min V,Ⅰ );

[0054] Yes, the maximum hourly rainfall in the IV threshold is max h,Ⅳ or the maximum maximum 3-hour rainfall I max 3h,Ⅳ or the maximum accumulated rainfall V max Ⅳ Updated to correspond to the maximum hourly rainfall I max h,<0.05 or the maximum maximum 3-hour rainfall I max 3h,<0.05 or the maximum accumulated rainfall V max <0.05 , and update the corresponding rainfall period; No, no update;

[0055] C6: Judge and optimize the rationality of the four-level critical rainfall threshold, and update the critical rainfall threshold of the corresponding waterlogging point after each alarm of the water depth monitoring equipment.

[0056] Preferably, the rationality of the four-level critical rainfall threshold is judged and optimized and adjusted, including:

[0057] D1: Extract the critical rainfall thresholds at each level, and determine whether each critical rainfall threshold satisfies the relationship of level IV threshold < level III threshold < level II threshold < level I threshold; if yes, do nothing; if no, that is, if the high level threshold is ≤ the low level threshold, jump to D2;

[0058] D2: Determine whether the critical rainfall threshold of Level IV threshold is greater than the thresholds of the previous levels; if yes, recalculate the critical rainfall threshold of Level IV threshold, and determine the size of the critical rainfall threshold of Level IV threshold and Level III threshold, Level II threshold, and Level I threshold respectively; if there is a threshold ≤ Level IV threshold, remove the threshold at that level, and query the minimum value of that level greater than Level IV threshold in the statistical table of waterlogging information at waterlogging points to update the critical rainfall threshold of that level; if no, further determine the size relationship among Level III threshold, Level II threshold, and Level I threshold, and readjust the threshold.

[0059] Specific:

[0060] First, count the number of non-null values ​​in level III, level II, and level I thresholds;

[0061] If the number of non-null values ​​is 0: that is, all are null values, no readjustment is required;

[0062] If the number of non-null values ​​is 1: no readjustment is required;

[0063] If the number of non-null values ​​is 2: determine the size of the low-level threshold and the high-level threshold. If the low-level threshold is ≥ the high-level threshold, remove the low-level threshold and make it null. If the low-level threshold is < the high-level threshold, no adjustment is required.

[0064] If the number of non-null values ​​is 3: determine the level of the maximum value. If the level I threshold is the maximum value, continue to determine whether level III is less than level II. If yes, no adjustment is required. If no, remove the level III threshold.

[0065] If the level II threshold is the maximum value, then continue to determine whether level III>level I; if yes, then remove the level I threshold; if no, then remove the level II threshold;

[0066] If the level III threshold is the maximum value, continue to determine whether level II is less than level I; if so, eliminate the level III threshold; if not, eliminate both the level III threshold and the level II threshold.

[0067] Preferably, the water depth monitoring device at the water accumulation point updates the critical rainfall threshold of the corresponding water accumulation point after each alarm, including:

[0068] E1: Extract the latest maximum hourly rainfall I obtained by this alarm from the water accumulation point water accumulation information sample database h new , Maximum 3-hour rainfall I 3h new and the cumulative rainfall V new The size of the stored threshold of the corresponding level; determine whether the stored threshold of the corresponding level is empty or greater than the latest value, that is:

[0069] I h,z new <I h,z min , or I 3h,z new <I 3h,z min , or V new <V z min ; Wherein, z is the classification code of each level, z can be Ⅳ, Ⅲ, Ⅱ or Ⅰ;

[0070] If yes, update the stored threshold of the corresponding level to the latest value, and record the rainfall period corresponding to the minimum value; if no, do nothing;

[0071] E2: Extract each critical rainfall threshold of the updated maximum hourly rainfall, maximum 3-hour rainfall and cumulative rainfall, and determine whether each critical rainfall threshold satisfies the size distribution relationship of Level IV threshold < Level III threshold < Level II threshold < Level I threshold; if yes, do not perform any operation; if no, if the latest critical rainfall threshold is not greater than a lower level threshold of the critical rainfall threshold, then the lower level threshold of the critical rainfall threshold is eliminated; if there is a critical rainfall threshold greater than a higher level threshold of the critical rainfall threshold, then the level threshold is not updated first and is restored to the original value or null value.

[0072] Preferably, the prediction of waterlogging time and risk based on the four-level threshold table of critical rainfall at waterlogging points includes:

[0073] F1: Start the prediction operation and mark the start time, obtain the actual rainfall data of each waterlogging point 3 hours before the start time, and obtain the upcoming forecast rainfall data of each waterlogging point from the forecast rainfall data at the location of the waterlogging point; among them, the actual rainfall data is obtained from the actual data of the rainfall station closest to the waterlogging point, and the forecast rainfall data is obtained from the weather forecast at the location of the waterlogging point; the upcoming forecast rainfall data is the forecast rainfall data within a certain period of time after the start time;

[0074] F2: Calculate the rainfall in the previous hour and the previous three hours at the start time based on the actual rainfall data of the waterlogging point, and compare them with the corresponding thresholds of each level in the four-level threshold table of critical rainfall of the waterlogging point, to determine whether the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours are both less than Level IV; if yes, send out the information that there is no risk of waterlogging at the current start time, and jump to F3; if no, take the highest level of the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours as the predicted waterlogging level, take the current start time as the predicted waterlogging time, send out the information that there is a risk of waterlogging at the current start time, and output the predicted waterlogging level and predicted waterlogging time;

[0075] F3: Extract the update time interval of the actual rainfall data and the forecast rainfall data; determine whether the update time interval of the actual rainfall data is less than or equal to the forecast rainfall data; if yes, mark the predictable time based on the start time and the update time interval of the forecast rainfall data; if no, jump to F4;

[0076] F4: Determine whether the rainfall in the three hours before the current start time is 0; if yes, still mark the predictable time based on the start time and the update time interval of the forecast rainfall data; if no, mark the predictable time based on the start time and the update time interval of the actual rainfall data;

[0077] F5: Extract predictable moments in sequence, calculate the rainfall in the previous hour and the previous three hours at the waterlogging point at the extracted predictable moment based on the actual rainfall data and forecast rainfall data of the waterlogging point, and compare them with the corresponding thresholds of each level in the four-level threshold table of critical rainfall of the waterlogging point, and judge whether the threshold levels of the 1-hour rainfall and the 3-hour rainfall are both less than Level IV; if yes, jump to F6; if no, take the highest level of the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours of the predictable moment as the predicted waterlogging level, take the current predictable moment as the predicted waterlogging moment, send out the information that there is a risk of waterlogging at the current predictable moment, and output the predicted waterlogging level and predicted waterlogging moment;

[0078] F6: Extract the next predictable moment and repeat F5 step until a predictable moment with waterlogging risk is found, then issue the predicted waterlogging level and predicted waterlogging time. If no predictable moment with waterlogging risk is found, stop judging when the predictable moment extraction is completed.

[0079] Compared with the prior art, the present invention has the following beneficial effects:

[0080] 1. The present invention dynamically tracks each rainfall, accumulates and obtains statistical data of water accumulation information at waterlogging points, takes this as sample data, obtains statistical data of rainfall periods when there is no waterlogging at waterlogging points, and uses this as the basis for identifying changes in the state of waterlogging points. A four-level critical rainfall threshold for waterlogging points is set, and three critical rainfall indicators are selected. The critical rainfall thresholds of each level and each indicator are extracted by the dynamic maximum value method, which solves the technical problems of being unable to timely identify waterlogging points when there are few sample data and changes in their state, as well as low accuracy in waterlogging prediction due to the limitations of traditional mechanism models. The present invention can more meticulously identify and distinguish the state of waterlogging points, dynamically obtain critical rainfall thresholds for waterlogging, and help improve the scientificity and effectiveness of urban waterlogging management, as well as the accuracy and real-time nature of predictions.

[0081] 2. The present invention proposes to use the maximum hourly rainfall, the maximum 3-hour rainfall and the accumulated rainfall as critical rainfall indicators, wherein the maximum hourly rainfall and the maximum 3-hour rainfall represent the rainfall intensity under different rainfall durations, and are suitable for a wide range of scenarios where waterlogging is caused by short-term heavy rainfall, while the accumulated rainfall is suitable for scenarios where waterlogging is caused by long-term rainfall and large accumulated rainfall. Therefore, the selected critical rainfall indicator can cover most rainfall waterlogging scenarios.

[0082] 3. The method for calculating and dynamically updating the four-level critical rainfall threshold value proposed in the present invention uses the statistics of water accumulation information at waterlogging points and the statistics of rainfall periods without waterlogging at waterlogging points as sample databases, and updates them in real time according to the dynamic tracking results of rainfall events after each alarm at the waterlogging points, thereby ensuring the reliability of the sample data; when obtaining the waterlogging information at the waterlogging points, the delay time of the "rainfall-surface runoff-pipeline confluence-ground waterlogging" process is taken into account by obtaining the critical water depth, critical rainfall moment, and critical rainfall duration, so that the calculation of the critical rainfall index is more in line with the actual physical process; the method for calculating the initial value of the four-level critical rainfall threshold value and dynamically adjusting the method repeatedly cross-compares the rainfall periods with waterlogging and the rainfall periods without waterlogging at the waterlogging points, and timely considers the changes in the state of the waterlogging points caused by human intervention, engineering rectification, etc., so that the calculation results of the critical rainfall threshold value are more accurate.

[0083] 4. The method proposed in the present invention realizes real-time and rolling prediction of waterlogging risk based on critical rainfall threshold, combines actual rainfall data and forecast rainfall data, and calculates the critical rainfall index value at each moment in the forecast period on a rolling basis. By comparing the dynamic evaluation results of the critical rainfall threshold, it can quickly predict whether the waterlogging point has waterlogging risk, including predicting the waterlogging level and predicting the waterlogging time. The prediction process has low computational complexity and high computational efficiency, and the prediction results are accurate and information-rich. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0085] Figure 1 It is a schematic diagram of the operation steps of the present invention;

[0086] Figure 2 A schematic diagram of the operation steps for calculating the rainfall index of the present invention;

[0087] Figure 3 A schematic diagram of the operation steps of obtaining the critical rainfall moment and critical rainfall duration of the water accumulation level in the present invention;

[0088] Figure 4 This is a schematic diagram of the rolling calculation of hourly rainfall in a fixed time window mode of the present invention. DETAILED DESCRIPTION

[0089] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0090] See also Figure 1-Figure 2 The first embodiment of the present invention provides a method for dynamically evaluating critical rainfall at a waterlogging point and predicting waterlogging risk, comprising:

[0091] Obtain the rainfall start time and rainfall end time of each rainfall, calculate the rainfall index for each rainfall, and store the rainfall start time, rainfall end time and rainfall index of each rainfall in the total database;

[0092] Establish a waterlogging point database, calculate and dynamically update the critical rainfall threshold based on the waterlogging point database to obtain a four-level threshold table of critical rainfall at the waterlogging point; wherein the waterlogging point database includes a waterlogging point waterlogging information sample database and a waterlogging point non-waterlogging rainfall period database;

[0093] The time and risk of waterlogging are predicted based on the four-level threshold table of critical rainfall at waterlogging points.

[0094] In this application, the rainfall start time and rainfall end time of each rainfall are obtained, including:

[0095] A1: Determine whether the water depth monitoring equipment at the water accumulation point has an alarm; if yes, record the time t0 when the monitoring alarm is generated and jump to A2; if no, do nothing;

[0096] A2: The nth hour before the monitoring alarm generation time t0 is recorded as t 0-n , traverse by time t 0-n The rainfall data from time to time t0 is used to determine whether a rainfall greater than V can be found in this traversal process. min rainfall data; yes, the first one greater than V min The collection time of rainfall data is marked as the rainfall start time t s , and jump to A3; No, stop analyzing the rainfall start time and rainfall end time corresponding to this alarm; where the value of n is determined according to the area size of the drainage unit, V min Indicates the minimum resolution for rainfall monitoring;

[0097] A3: t s The time is the first starting time, and the rainfall data from the latest first starting time to n hours before the first starting time are traversed again. When a rainfall greater than V can be found in this traversal process min When the rainfall data is available, use the first value greater than V min The collection time of rainfall data is relative to the rainfall start time t s Make updates;

[0098] A4: Repeat step A3 until no rainfall greater than V is found in the rainfall data from the latest first starting point to n hours before the first starting point. min When the rainfall data is output, the rainfall start time t s ;

[0099] A5: The time after the monitoring alarm generation time t0 is recorded as t 0+1 , traverse from time t0 to t 0+1 The rainfall data at each moment is used to determine whether a rainfall greater than V can be found during this traversal. min rainfall data; yes, the last one greater than V min The collection time of rainfall data is marked as the end time of rainfall t e , with t e The time is the second starting time; otherwise, t 0+1 The time is the second starting time;

[0100] A6: traverse the rainfall data from the latest second starting point to 1 hour after the second starting point again; when a rainfall greater than V can be found in this traversal process min When the rainfall data is greater than V, use the last minThe collection time of rainfall data is relative to the rainfall end time t e Update and use the latest rainfall end time t e Update the second starting point time; when no value greater than V is found in this traversal process min When the rainfall data is obtained, the second starting point time is updated with the time 1 hour after the second starting point time used in this traversal;

[0101] A7: Repeat step A6 until the latest rainfall ends at time t e As the starting point, no value greater than V can be found in the next n hours. min When the rainfall data is obtained; determine the rainfall end time t e Is it empty? If yes, the monitoring alarm generation time t0 is used as the rainfall end time t e Output; No, output the latest rainfall end time t e .

[0102] It should be noted that the data used in this step to obtain the start time and end time of each rainfall is the data recorded by the rainfall station that is closest to the water accumulation point in a straight line.

[0103] It should be noted that the value of n is determined according to the area of ​​the drainage unit, generally the draining time of the drainage unit; wherein the drainage unit is obtained by artificial selection.

[0104] It should be noted that the traversal process used in the present invention is to traverse from front to back in time.

[0105] It should be noted that, by traversing t 0-n In the rainfall time series data from time t to time t0, the time points of the traversal process do not include t 0-n Time, including time t0.

[0106] It should be noted that in step A2, when no value greater than V can be found in this traversal process, min When there is rainfall data, that is, there is no obvious rainfall in the previous n hours, the water accumulation point alarm may be a false alarm, and the rainfall start time and rainfall end time corresponding to this alarm will no longer be analyzed.

[0107] It should be noted that, from time t0 to t 0+1 In the momentary rainfall data, the time points of the traversal process include t0 and t 0+1 time.

[0108] It should be noted that, when determining the rainfall end time t e Whether it is in the air, when in the traversal process after the monitoring alarm is generated at time t0, no value greater than Vmin When the rainfall data is empty; when the traversal process after the monitoring alarm is generated at time t0, it is possible to find a rainfall data greater than V min The rainfall data is not null.

[0109] This application calculates rainfall indicators for each rainfall event, including:

[0110] B1: Extract the rainfall start time t of this rainfall s and the rainfall end time t e , judge t e -t s Is it greater than 3 hours? If yes, calculate the rainfall every 3 hours using the sliding time window and compare to get t s Time t e Maximum 3-hour rainfall between and 3h ; No, jump to B2;

[0111] B2: Determine t e -t s Is it greater than 2 hours and 30 minutes? If yes, use t s Time t e The accumulated rainfall between and is taken as the maximum 3-hour rainfall I 3h ; No, set the maximum 3-hour rainfall I 3h Set to null value;

[0112] B3: Extract the rainfall start time t of this rainfall s and the rainfall end time t e , judge t e -t s Is it greater than 1 hour? If yes, calculate the rainfall every 1 hour using the sliding time window and compare to get t s Time t e Maximum hourly rainfall between and h ; No, jump to B4;

[0113] B4: Determine t e -t s Is it greater than 40 minutes? Yes, in t s Time t e The accumulated rainfall between and is taken as the maximum hourly rainfall I h ; No, set the maximum hourly rainfall I h Set to null value.

[0114] It should be noted that the sliding time window in calculating the rainfall at every 3-hour time interval using the sliding time window and the sliding time window in calculating the rainfall at every 1-hour time interval using the sliding time window can be described in detail as follows:

[0115] Get the data collection period ΔT of the rainfall station corresponding to the water accumulation point, and determine the length of the time window as 1+60 / ΔT. For example, the calculation of rainfall every 1-hour time interval using a sliding time window can be illustrated as follows: data collection period ΔT = 5 minutes, then the length of the time window is 1+60 / ΔT = 1+60 / 5 = 13. For details, please refer to Figure 4 ; The unit of ΔT is minutes.

[0116] In this application, a waterlogging point database is established, including:

[0117] Extract all rainfall data stored in the database corresponding to the current waterlogging point, and extract the rainfall period t of each rainfall data i s ~t i e , extract each rainfall period t i s ~t i e The maximum water depth h of the corresponding water point i max ; where i is the recording order of rainfall data, rainfall period t i s ~t i e The time range is from time t i s At time t i e ;

[0118] The maximum water depth h i max The rainfall period of not less than 0.05 meters i s ~t i e Mark as target time period, extract the maximum water depth h in each target time period i max The waterlogging level is located, and the critical rainfall time and critical rainfall duration of the waterlogging level are obtained by interpolation method; among them, the waterlogging level includes level IV waterlogging level: 0<h max ≤h 4,max , Grade III water accumulation level: h 4,max <h max ≤h 3,max , Level II water accumulation level: h 3,max <h max ≤h 2,max , Grade I waterlogging level: h 2,max <h max ;

[0119] The maximum water depth h in the target time period corresponding to each waterlogging point is obtained. i max , water accumulation level, rainfall start time t i s , Rainfall end time t i e , critical rainfall time, critical rainfall duration, and the maximum hourly rainfall obtained according to the critical rainfall duration I h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i The water accumulation information is recorded in a water accumulation point water accumulation information statistical table to form a water accumulation point water accumulation information sample database;

[0120] The maximum water depth h i max The rainfall period is less than 0.05 meters. i s ~t i e Mark as non-target time period, and calculate the maximum hourly rainfall I in each non-target time period according to steps A1 to A7 and steps B1 to B4. h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i ＜0.05 ;

[0121] The maximum hourly rainfall I h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i ＜0.05 The data are recorded in the statistical table of rainfall periods without water accumulation at water accumulation points, forming a database of rainfall periods without water accumulation at water accumulation points.

[0122] It should be noted that at time t i s Indicates the rainfall start time of the rainfall data with recording order i; time t i e Indicates the rainfall end time of the rainfall data with recording order i.

[0123] It should be noted that the maximum hourly rainfall I obtained according to the critical rainfall duration h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i The water accumulation information is the critical rainfall duration in the target time period.

[0124] It should be noted that the maximum water depth h i max It is measured by water depth monitoring equipment at the water accumulation point.

[0125] It should be noted that the ranking of the Level IV waterlogging level, Level III waterlogging level, Level II waterlogging level, and Level I waterlogging level is based on the depth of the water, and the ranking from high to low is: Level I waterlogging level, Level II waterlogging level, Level III waterlogging level, and Level IV waterlogging level.

[0126] See also Figure 3 In this application, the interpolation method is used to obtain the critical rainfall moment and critical rainfall duration of the water accumulation level, including:

[0127] Extract the waterlogging level and determine whether it is level IV. If yes, use interpolation to obtain the maximum waterlogging depth h in this record. i max The first time when the rainfall is equal to 0.05 meters is marked as the critical rainfall moment of Grade IV flooding level t i 0.05 ; No, use the interpolation method to obtain the moment when the water depth of the waterlogging point reaches the minimum value of the waterlogging level, and mark the moment as the critical rainfall moment of the waterlogging level; among them, the critical rainfall moment of level III waterlogging level is marked as t i (h4,max-Δh3,cr) , the critical rainfall moment of level II flooding is marked as t i (h3,max-Δh2,cr) , the critical rainfall moment of level I flooding is marked as t i (h2,max-Δh1,cr) ;

[0128] Get the time interval of the critical rainfall moment [t j-1 ,t j ], and the time interval [t j-1 ,t j ] Total rainfall V j , calculate t using interpolation j-1 The rainfall Vz in the time period between the critical rainfall moment and the critical rainfall moment; where the time interval [t j-1 ,t j ] is the collection time of rainfall data, and z in rainfall Vz is the classification number of waterlogging level;

[0129] Calculate the rainfall start time t i s Critical rainfall time to level IV flooding level t i 0.05 The critical rainfall duration Δt i 4. Calculate the rainfall start time t i s Critical rainfall time to level III flooding level t i (h4,max-Δh3,cr) The critical rainfall duration Δt i 3. Calculate the rainfall start time ti s Critical rainfall time to level II flooding level t i (h3,max-Δh2,cr) The critical rainfall duration Δt i 2. Calculate the rainfall start time t i s The critical rainfall moment for level I flooding is marked as t i (h2,max-Δh1,cr) The critical rainfall duration Δt i 1.

[0130] It should be noted that the critical rainfall moment of the waterlogging level is the moment when the water depth at the waterlogging point reaches the minimum value of the waterlogging level range.

[0131] It should be noted that interpolation is a mathematical method that uses known data points to estimate unknown data points. In the present invention, interpolation is used to obtain the critical rainfall time t in this record. i 0.05 ,t i (h4,max-Δh3,cr) ,t i (h3,max-Δh2,cr) and t i (h2,max-Δh1,cr) The critical rainfall time t of level IV flooding can be i 0.05 Take an example to understand:

[0132]

[0133] Among them, t i ＜0.05 and h i ＜0.05 is the collection time value and water depth less than 0.05 m in the water depth monitoring data; t i ＞0.05 and h i ＞0.05 It is the collection time value and water depth greater than 0.05 meters in the water depth monitoring data.

[0134] It should be noted that the time interval [t j-1 ,t j ] is the collection time of rainfall data. An example can be given as follows: if the time of a critical rainfall moment is 14:02 on December 20, 2024, and the collection period of rainfall data is 5 minutes, and 14:02 on December 20, 2024 is in the collection interval of rainfall data from 14:00 on December 20, 2024 to 14:05 on December 20, 2024, the time interval of the critical rainfall moment is [2024-12-20-14:00, 2024-12-20-14:05].

[0135] It should be noted that in the rainfall Vz, z is the classification code of the waterlogging level. When z is Ⅳ, it means the rainfall Vt of the Ⅳ waterlogging level. i 0.05 When z is Ⅲ, it indicates the rainfall Vt of level Ⅲ flooding level i (h4,max-Δh3,cr) When z is Ⅱ, it indicates the rainfall Vt of level Ⅱ waterlogging level i (h3,max-Δh2,cr) When z is Ⅰ, it indicates the rainfall Vt of level Ⅰ waterlogging level i (h2,max-Δh1,cr) .

[0136] It should be noted that the interpolation method is used to calculate t j-1 The rainfall Vz in the time period between the critical rainfall time and the critical rainfall time can be illustrated as follows: i 0.05 For example, interpolation obtains t j-1 to i 0.05 The rainfall Vt i 0.05 :

[0137]

[0138] It should be noted that after obtaining the critical rainfall duration corresponding to each critical rainfall moment, the specific steps for calculating the critical rainfall index value in each critical rainfall duration are as follows:

[0139] If 2 hours and 30 minutes ≤ critical rainfall duration ≤ 6 hours, then calculate the critical rainfall index values ​​within the critical rainfall duration according to steps A1 to A7 and steps B1 to B4; wherein the critical rainfall index values ​​include the maximum hourly rainfall I h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i ;

[0140] If 40 minutes ≤ critical rainfall duration < 2 hours and 30 minutes, then only steps A1 to A7 and steps B1 to B4 are used to calculate the maximum hourly rainfall I for the critical rainfall duration. h ;

[0141] If the critical rainfall duration is less than 40 minutes, the critical rainfall index values ​​will not be calculated.

[0142] It should be noted that the statistical table of water accumulation information at water accumulation points is shown in Table 1.

[0143]

[0144] Table 1 Statistics of waterlogging information at waterlogging points It should be noted that the statistics of rainfall periods without waterlogging at waterlogging points are shown in Table 2.

[0145]

[0146] Table 2 Statistics of rainfall periods with or without waterlogging at waterlogging points

[0147] In this application, the critical rainfall threshold is calculated and dynamically updated to obtain a four-level threshold table of critical rainfall at waterlogging points, including:

[0148] According to the latest waterlogging information sample database, the rainfall periods with the same waterlogging level are compared and the minimum maximum hourly rainfall I min h,z , minimum maximum 3-hour rainfall I min 3h,z and the minimum accumulated rainfall V min z As the initial value of the critical rainfall threshold of the corresponding level, and record the rainfall period (t i s ~t i e ) min w,z , the minimum maximum hourly rainfall I in each waterlogging level at the waterlogging point min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min And the rainfall period corresponding to each minimum value (t i s ~t i e ) min w,z Recorded in the four-level threshold table of critical rainfall at waterlogging points; among them, rainfall period (t i s ~t i e ) min w,z Where w is the classification number of rainfall, and z is the classification number of waterlogging level;

[0149] The critical rainfall thresholds are dynamically updated to obtain a four-level threshold table of critical rainfall at waterlogging points.

[0150] In this application, the critical rainfall threshold is dynamically updated to obtain a four-level threshold table of critical rainfall at waterlogging points, including:

[0151] C1: Determine whether the IV-level threshold stored in the IV-level threshold table of critical rainfall at waterlogging points is not empty; if yes, jump to C2; if no, jump to C5; among which, non-empty means there is a specific value;

[0152] C2: Extract the rainfall period corresponding to the stored IV level threshold, mark all rainfall periods after the rainfall period corresponding to the stored IV level threshold as reference rainfall periods in the latest rainfall period database without water accumulation, and count the maximum value of the maximum hourly rainfall in the reference rainfall period. h lat_max and the second largest value I h lat_semax , the maximum value of the maximum rainfall in 3 hours I 3h lat_max and the second largest value I 3h lat_semax , the maximum value of the accumulated rainfall V lat_max and the next largest value V lat_semax ;

[0153] C3: If the maximum value and the second maximum value are ≥ K times the existing IV level threshold of the corresponding indicator, that is:

[0154] I h lat_max ,I h lat_semax ≥K×I min h,Ⅳ ;

[0155] or I 3h lat_max ,I 3h lat_semax ≥K×I min 3h,Ⅳ ;

[0156] or V lat_max 、V lat_semax ≥K×V min Ⅳ ;

[0157] Among them, K is a proportional adjustment coefficient greater than 1;

[0158] C4: Determine whether the maximum value and the second maximum value in the three formulas of C3 are ≥ K times the existing IV-level threshold of the corresponding indicator. If yes, mark the earliest rainfall period in the rainfall period corresponding to the established formula as the earliest rainfall period. In the waterlogging point waterlogging information sample database, if the minimum maximum hourly rainfall I with waterlogging level of IV after the earliest rainfall period can be found, min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min, the corresponding minimum value is used as the latest IV level threshold; if the minimum maximum hourly rainfall I with water accumulation level of IV after the earliest rainfall period cannot be found, min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min , then directly use the maximum value of the maximum hourly rainfall in the reference rainfall period I h lat_max , the maximum value of the maximum rainfall in 3 hours I 3h lat_max , the maximum value of the accumulated rainfall V lat_max As the latest level IV threshold;

[0159] No, do not update the existing Level IV threshold;

[0160] C5: Query the latest database of rainfall periods with or without water accumulation, and extract the maximum hourly rainfall I max h,<0.05 , Maximum 3-hour rainfall I max 3h,<0.05 and the maximum accumulated rainfall V max <0.05 , and the rainfall period corresponding to each maximum value (t i s ~t i e ) max w,<0.05 ; Determine whether the maximum value is not less than the minimum value of the threshold corresponding to each level, and the rainfall period corresponding to the maximum value is greater than the rainfall period corresponding to the minimum value of the threshold, that is:

[0161] I max h,<0.05 ≥min(I min h,Ⅲ ,I min h,Ⅱ ,I min h,Ⅰ ), and (t i s ~t i e ) max h,<0.05 >min((t i s ~t i e ) min h,Ⅲ ,(t i s ~t i e ) minh,Ⅱ ,(t i s ~t i e ) min h,Ⅰ );

[0162] or I max 3h,<0.05 ≥min(I min 3h,Ⅲ ,I min 3h,Ⅱ ,I min 3h,Ⅰ ), and (t i s ~t i e ) max 3h,<0.05 >min((t i s ~t i e ) min 3h,Ⅲ ,(t i s ~t i e ) min 3h,Ⅱ ,(t i s ~t i e ) min 3h,Ⅰ );

[0163] or V max <0.05 ≥min(V min Ⅲ ,V min Ⅱ ,V min Ⅰ ), and (t i s ~t i e ) max V,<0.05 >min((t i s ~t i e ) min V,Ⅲ ,(t i s ~t i e ) min V,Ⅱ ,(t is ~t i e ) min V,Ⅰ );

[0164] Yes, the maximum hourly rainfall in the IV threshold is max h,Ⅳ or the maximum maximum 3-hour rainfall I max 3h,Ⅳ or the maximum accumulated rainfall V max Ⅳ Updated to correspond to the maximum hourly rainfall I max h,<0.05 or the maximum maximum 3-hour rainfall I max 3h,<0.05 or the maximum accumulated rainfall V max <0.05 , and update the corresponding rainfall period; No, no update;

[0165] C6: Judge and optimize the rationality of the four-level critical rainfall threshold, and update the critical rainfall threshold of the corresponding waterlogging point after each alarm of the water depth monitoring equipment.

[0166] It is worth noting that the method for calculating and dynamically updating the four-level critical rainfall threshold value proposed in the present invention uses the statistics of water accumulation information at waterlogging points and the statistics of rainfall periods without waterlogging at waterlogging points as the sample database, and updates it in real time according to the dynamic tracking results of rainfall events after each alarm at the waterlogging points, thereby ensuring the reliability of the sample data; when obtaining the waterlogging information at the waterlogging points, the delay time of the "rainfall-surface runoff-pipeline confluence-ground waterlogging" process is taken into account by obtaining the critical water depth, critical rainfall moment, and critical rainfall duration, so that the calculation of the critical rainfall index is more in line with the actual physical process; the present invention proposes a method for calculating the initial values ​​and dynamically adjusting the four-level critical rainfall threshold values, repeatedly cross-comparing the rainfall periods with waterlogging and the rainfall periods without waterlogging at the waterlogging points, and timely considering the changes in the state of the waterlogging points caused by human intervention, engineering rectification, etc., so that the calculation results of the critical rainfall threshold values ​​are more accurate.

[0167] It should be noted that the rainfall period (t i s ~t i e ) min w,z The specific description of where w is the classification number of rainfall and z is the classification number of waterlogging level is:

[0168] When w is h, it indicates the minimum maximum hourly rainfall I min h,z The rainfall period (t i s~t i e ) min h,z When w is 3h, it indicates the minimum maximum 3-hour rainfall I min 3h,z The rainfall period (t i s ~t i e ) min 3h,z , when w is V, it indicates the minimum accumulated rainfall V min z The rainfall period (t i s ~t i e ) min V,z ;

[0169] When z is Ⅳ, it indicates the rainfall period of level Ⅳ threshold (t i s ~t i e ) min w,Ⅳ , the minimum maximum hourly rainfall in the IV level threshold min h,Ⅳ , minimum maximum 3-hour rainfall I min 3h,Ⅳ and the minimum accumulated rainfall V min Ⅳ ; When z is Ⅲ, it indicates the rainfall period of the Ⅲ level threshold (t i s ~t i e ) min w,Ⅲ , the minimum maximum hourly rainfall in the level III threshold min h,Ⅲ , minimum maximum 3-hour rainfall I min 3h,Ⅲ and the minimum accumulated rainfall V min Ⅲ ; When z is Ⅱ, it indicates the rainfall period of level Ⅱ threshold (t i s ~t i e ) min w,z , the minimum maximum hourly rainfall in the level II threshold min h,Ⅱ , minimum maximum 3-hour rainfall I min 3h,Ⅱ and the minimum accumulated rainfall V min Ⅱ; When z is Ⅰ, it indicates the rainfall period of level Ⅰ threshold (t i s ~t i e ) min w,z , the minimum maximum hourly rainfall in the level I threshold min h,Ⅰ , minimum maximum 3-hour rainfall I min 3h,Ⅰ and the minimum accumulated rainfall V min Ⅰ .

[0170] It should be noted that the four-level threshold table of critical rainfall at waterlogging points is shown in Table 3.

[0171]

[0172]

[0173] Table 3 Four-level thresholds of critical rainfall at waterlogging points

[0174] It should be noted that among the stored IV thresholds in the table of four thresholds for judging critical rainfall at waterlogging points, the stored IV thresholds represent the minimum maximum hourly rainfall I corresponding to the IV waterlogging level. min h , Maximum 3-hour rainfall I min 3h and the accumulated rainfall V min .

[0175] It should be noted that in step C4, the three formulas in the three formulas in C3 are judged to be: h lat_max ,I h lat_semax ≥K×I min h,Ⅳ ; or 3h lat_max ,I 3h lat_semax ≥K×I min 3h,Ⅳ ; or V lat_max 、V lat_semax ≥K×V min Ⅳ ;

[0176] Whether the maximum value and the second maximum value are ≥ K times the existing IV-level threshold of the corresponding indicator is established: I h lat_max ,I h lat_semax ≥K×I min h,Ⅳ or I 3hlat_max ,I 3h lat_semax ≥K×I min 3h,Ⅳ or V lat_max 、V lat_semax ≥K×V min Ⅳ , if one of the three equations holds true, it means it is true.

[0177] It should be noted that in the rainfall period corresponding to each maximum value (t i s ~t i e ) max w,<0.05 When w is h, it indicates the maximum hourly rainfall I max h,<0.05 The corresponding rainfall period (t i s ~t i e ) max h,<0.05 , when w is 3h, it means the maximum rainfall in 3 hours I max 3h,<0.05 The corresponding rainfall period (t i s ~t i e ) max 3h,<0.05 , when w is V, it indicates the maximum accumulated rainfall V max <0.05 The corresponding rainfall period (t i s ~t i e ) max V,<0.05 .

[0178] In this application, the rationality of the four-level critical rainfall threshold is judged and optimized, including:

[0179] D1: Extract the critical rainfall thresholds at each level, and determine whether each critical rainfall threshold satisfies the relationship of level IV threshold < level III threshold < level II threshold < level I threshold; if yes, do nothing; if no, that is, if the high level threshold is ≤ the low level threshold, jump to D2;

[0180] D2: Determine whether the critical rainfall threshold of Level IV threshold is greater than the thresholds of the previous levels; if yes, recalculate the critical rainfall threshold of Level IV threshold, and determine the size of the critical rainfall threshold of Level IV threshold and Level III threshold, Level II threshold, and Level I threshold respectively; if there is a threshold ≤ Level IV threshold, remove the threshold at that level, and query the minimum value of that level greater than Level IV threshold in the statistical table of waterlogging information at waterlogging points to update the critical rainfall threshold of that level; if no, further determine the size relationship among Level III threshold, Level II threshold, and Level I threshold, and readjust the threshold.

[0181] Specific:

[0182] First, count the number of non-null values ​​in level III, level II, and level I thresholds;

[0183] If the number of non-null values ​​is 0: that is, all are null values, no readjustment is required;

[0184] If the number of non-null values ​​is 1: no readjustment is required;

[0185] If the number of non-null values ​​is 2: determine the size of the low-level threshold and the high-level threshold. If the low-level threshold is ≥ the high-level threshold, remove the low-level threshold and make it null. If the low-level threshold is < the high-level threshold, no adjustment is required.

[0186] If the number of non-null values ​​is 3: determine the level of the maximum value. If the level I threshold is the maximum value, continue to determine whether level III is less than level II. If yes, no adjustment is required. If no, remove the level III threshold.

[0187] If the level II threshold is the maximum value, then continue to determine whether level III>level I; if yes, then remove the level I threshold; if no, then remove the level II threshold;

[0188] If the level III threshold is the maximum value, continue to determine whether level II is less than level I; if so, eliminate the level III threshold; if not, eliminate both the level III threshold and the level II threshold.

[0189] It is worth noting that due to the uneven spatial distribution of rainfall and the certain distance between the rain gauge and the flood-prone point, the monitored rainfall of the rain gauge may not truly reflect the actual rainfall in the flood-prone point, and the state of the waterlogging point may change. Therefore, the four-level thresholds initially obtained by the above judgment are likely to not meet the size relationship of level IV threshold < level III threshold < level II threshold < level I threshold. After obtaining the final critical rainfall threshold, for each critical rainfall index of the maximum hourly rainfall, the maximum 3-hour rainfall and the cumulative rainfall, it is necessary to judge whether the size relationship of level IV threshold < level III threshold < level II threshold < level I threshold is met. If it is satisfied, it means that the distribution of each level threshold is reasonable. If it is not satisfied, that is, the situation of high-level threshold ≤ low-level threshold appears, then it is necessary to re-judge and adjust the threshold. The present invention can avoid ensuring that the rainfall assessment can more accurately reflect the actual rainfall situation of the flood-prone point by making reasonable judgments and optimizing adjustments to the thresholds at each level, which helps to reduce false alarms and missed reports and improve the accuracy and reliability of the analysis.

[0190] It should be noted that the rank order of level IV threshold, level III threshold, level II threshold, and level I threshold is level I threshold, level II threshold, level III threshold, and level IV threshold from front to back.

[0191] In this application, the water depth monitoring device at the waterlogging point updates the critical rainfall threshold of the corresponding waterlogging point after each alarm, including:

[0192] E1: Extract the latest maximum hourly rainfall I obtained by this alarm from the water accumulation point water accumulation information sample database h new , Maximum 3-hour rainfall I 3h new and the cumulative rainfall V new The size of the stored threshold of the corresponding level; determine whether the stored threshold of the corresponding level is empty or greater than the latest value, that is:

[0193] I h,z new <I h,z min , or I 3h,z new <I 3h,z min , or V new <V z min ; Wherein, z is the classification code of each level, z can be Ⅳ, Ⅲ, Ⅱ or Ⅰ;

[0194] If yes, update the stored threshold of the corresponding level to the latest value, and record the rainfall period corresponding to the minimum value; if no, do nothing;

[0195] E2: Extract each critical rainfall threshold of the updated maximum hourly rainfall, maximum 3-hour rainfall and cumulative rainfall, and determine whether each critical rainfall threshold satisfies the size distribution relationship of Level IV threshold < Level III threshold < Level II threshold < Level I threshold; if yes, do not perform any operation; if no, if the latest critical rainfall threshold is not greater than a lower level threshold of the critical rainfall threshold, then the lower level threshold of the critical rainfall threshold is eliminated; if there is a critical rainfall threshold greater than a higher level threshold of the critical rainfall threshold, then the level threshold is not updated first and is restored to the original value or null value.

[0196] It should be noted that after each alarm, the water depth monitoring equipment at the water accumulation point has calculated and counted the latest rainfall period and water accumulation information, including the water accumulation level L new , rainfall period (t i s ~t i e ) new , the maximum hourly rainfall within the critical rainfall duration I h new , Maximum 3-hour rainfall I 3h new and the cumulative rainfall V new And other indicator values ​​have been saved in the waterlogging point waterlogging information sample database.

[0197] It should be noted that if the latest critical rainfall threshold is not greater than the lower level threshold of the critical rainfall threshold, the lower level threshold of the critical rainfall threshold is removed; if the critical rainfall threshold is greater than the higher level threshold of the critical rainfall threshold, the level threshold is not updated first and restored to the original value or null value. It can be specifically described as follows:

[0198] If there is a latest critical rainfall threshold that is not greater than the lower level threshold of the critical rainfall threshold, that is, I w,z new ≤I w,y1 min , then the lower level threshold of the critical rainfall threshold is eliminated. If there is a higher level threshold whose critical rainfall threshold is greater than the critical rainfall threshold, that is, I w,z new >I w,y1 min Then the level threshold is not updated first and restored to the original value or null value; where w is the classification label of rainfall, z<y1, z, y1 and y2 are the classification labels of the four levels of Ⅳ, Ⅲ, Ⅱ or Ⅰ;

[0199] It should be noted that at the critical rainfall threshold I w,z new and lower level threshold I w,y1 min

[0200] When w is h, it indicates the critical rainfall threshold I corresponding to the maximum hour. h,z new and lower level threshold I h,y1 min When w is 3h, it means the critical rainfall threshold I corresponding to the maximum 3 hours 3h,z new and lower level threshold I 3h,y1 min , when w is V, it means the critical rainfall threshold V corresponding to the accumulated rainfall new and the lower level threshold V z min .

[0201] This application predicts the time and risk of flooding based on the four-level threshold table of critical rainfall at flooding points, including:

[0202] F1: Start the prediction operation and mark the start time, obtain the actual rainfall data of each waterlogging point 3 hours before the start time, and obtain the upcoming forecast rainfall data of each waterlogging point from the forecast rainfall data at the location of the waterlogging point; among them, the actual rainfall data is obtained from the actual data of the rainfall station closest to the waterlogging point, and the forecast rainfall data is obtained from the weather forecast at the location of the waterlogging point; the upcoming forecast rainfall data is the forecast rainfall data within a certain period of time after the start time;

[0203] F2: Calculate the rainfall of the waterlogging point in the previous hour and the previous three hours based on the actual rainfall data of the waterlogging point at the start time, and compare them with the corresponding thresholds of each level in the four-level threshold table of critical rainfall of the waterlogging point, to determine whether the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours are both less than Level IV; if yes, send out the information that there is no risk of waterlogging at the current start time, and jump to F3; if no, take the highest level of the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours as the predicted waterlogging level, take the current start time as the predicted waterlogging time, send out the information that there is a risk of waterlogging at the current start time, and output the predicted waterlogging level and predicted waterlogging time;

[0204] F3: Extract the update time interval of the actual rainfall data and the forecast rainfall data; determine whether the update time interval of the actual rainfall data is less than or equal to the forecast rainfall data; if yes, mark the predictable time based on the start time and the update time interval of the forecast rainfall data; if no, jump to F4;

[0205] F4: Determine whether the rainfall in the three hours before the current start time is 0; if yes, still mark the predictable time based on the start time and the update time interval of the forecast rainfall data; if no, mark the predictable time based on the start time and the update time interval of the actual rainfall data;

[0206] F5: Extract predictable moments in sequence, calculate the rainfall in the previous hour and the previous three hours at the extracted predictable moment based on the actual rainfall data and forecast rainfall data of the waterlogging point, and compare them with the corresponding thresholds of each level in the four-level threshold table of critical rainfall of the waterlogging point, and judge whether the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours are both less than Level IV; if yes, jump to F6; if no, take the highest level of the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours of the predictable moment as the predicted waterlogging level, take the current predictable moment as the predicted waterlogging moment, send out the information that there is a risk of waterlogging at the current predictable moment, and output the predicted waterlogging level and predicted waterlogging moment;

[0207] F6: Extract the next predictable moment and repeat F5 step until a predictable moment with waterlogging risk is found, then issue the predicted waterlogging level and predicted waterlogging time. If no predictable moment with waterlogging risk is found, stop judging when the predictable moment extraction is completed.

[0208] It is worth noting that the critical rainfall threshold level classification standard proposed in the present invention sets four-level thresholds of critical rainfall for waterlogging according to the severity level of the maximum waterlogging depth of the waterlogging point, which correspond to each other. In specific implementation, the severity level of the maximum waterlogging depth of the waterlogging point can be set according to the current national, industry and local standards, so that the critical rainfall threshold has strong applicability and can be strictly used for accurate prediction and early warning of different waterlogging levels at waterlogging points; the present invention proposes to use the maximum hourly rainfall, the maximum 3-hour rainfall and the accumulated rainfall as critical rainfall indicators, among which the maximum hourly rainfall and the maximum 3-hour rainfall represent the rainfall intensity under different rainfall durations, and are suitable for a wide range of scenarios of urban waterlogging caused by short-term heavy rainfall, and the accumulated rainfall is suitable for scenarios of urban waterlogging caused by long-term rainfall and large accumulated rainfall. Therefore, the selected critical rainfall indicator can cover most rainfall urban waterlogging scenarios.

[0209] It is worth noting that the method proposed in the present invention for achieving real-time, rolling prediction of waterlogging risks based on critical rainfall thresholds, combines actual rainfall data with forecast rainfall data, and rolls over the critical rainfall index value at each moment in the forecast period to compare the dynamic evaluation results of the critical rainfall thresholds. This can quickly predict whether a waterlogging point has a risk of waterlogging, including predicting the level of waterlogging and the time of waterlogging. The prediction process requires little computation and has high efficiency, and the prediction results are accurate and information-rich.

[0210] It should be noted that when the prediction operation is started, in order to ensure that the forecast rainfall data has been updated in a timely manner, the algorithm start time may be delayed by a certain time interval, such as 3 to 5 minutes, compared to the forecast rainfall data update time.

[0211] It should be noted that when obtaining rainfall data and subsequent predictions, the delay time of algorithm startup may not be considered. If the delay time interval is 3 minutes and the startup time is 08:03, 08:00 is used as the startup time to obtain the actual rainfall data from 05:00 to 08:00 and the forecast rainfall data from 08:00 to 11:00; in this embodiment, the forecast rainfall data is the forecast rainfall data within 3 hours after the startup time.

[0212] It should be noted that the calculation of the rainfall in the previous hour and the previous three hours at the waterlogging point based on the actual rainfall data of the waterlogging point can be illustrated as follows: if the starting time is 08:03, the rainfall in the previous hour from 07:00 to 08:00 is calculated, and the rainfall in the previous three hours from 05:00 to 08:00 is calculated.

[0213] It should be noted that the predictable time can be marked based on the start time and the update time interval of the forecast rainfall data as follows:

[0214] If the start time is 08:00 and the update interval of the forecast rainfall data is 10 minutes, then 08:10, 08:20, 08:30, ..., 11:00 are the predictable times after the start time.

[0215] It should be noted that the predictable time is marked based on the start time and the update time interval of the real-time rainfall data, which can be illustrated as follows:

[0216] The current start time is 08:00, and the time interval of real-time rainfall is 15 minutes, so 08:15, 08:30, 08:45, 09:00, ..., 11:00 are the predictable times after the start time.

[0217] It should be noted that extracting the predictable moments in sequence means extracting them from the beginning to the end in time.

[0218] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0219] Working principle of the present invention:

[0220] The rainfall start time and rainfall end time of each rainfall are obtained, and the rainfall index is calculated for each rainfall, and the rainfall start time, rainfall end time and rainfall index of each rainfall are stored in the total database; a waterlogging point database is established, including a waterlogging point waterlogging information sample database and a waterlogging point non-waterlogging rainfall period database; based on the waterlogging point database, including the waterlogging point waterlogging information sample database and the waterlogging point non-waterlogging rainfall period database, the critical rainfall threshold is calculated and dynamically updated to obtain a four-level threshold table of critical rainfall for waterlogging points.

[0221] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk, characterized in that: Obtain the rainfall start time and rainfall end time of each rainfall, calculate the rainfall index for each rainfall, and store the rainfall start time, rainfall end time and rainfall index of each rainfall in the total database; Establish a waterlogging point database, calculate and dynamically update the critical rainfall threshold based on the waterlogging point database to obtain a four-level threshold table of critical rainfall at the waterlogging point; wherein the waterlogging point database includes a waterlogging point waterlogging information sample database and a waterlogging point non-waterlogging rainfall period database; The time and risk of waterlogging are predicted based on the four-level threshold table of critical rainfall at waterlogging points.

2. According to claim 1, a method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk is characterized in that: The step of obtaining the rainfall start time and rainfall end time of each rainfall includes: A1: Determine whether the water depth monitoring equipment at the water accumulation point has an alarm; if yes, record the time t0 when the monitoring alarm is generated and jump to A2; if no, do nothing; A2: The nth hour before the monitoring alarm generation time t0 is recorded as t 0-n , traverse by time t 0-n The rainfall data from time to time t0 is used to determine whether a rainfall greater than V can be found in this traversal process. min rainfall data; yes, the first one greater than V min The collection time of rainfall data is marked as the rainfall start time t s , and jump to A3; No, stop analyzing the rainfall start time and rainfall end time corresponding to this alarm; where the value of n is determined according to the area size of the drainage unit, V min Indicates the minimum resolution for rainfall monitoring; A3: t s The time is the first starting time, and the rainfall data from the latest first starting time to n hours before the first starting time are traversed again. When a rainfall greater than V can be found in this traversal process min When the rainfall data is available, use the first value greater than V min The collection time of rainfall data is relative to the rainfall start time t s Make updates; A4: Repeat step A3 until no rainfall greater than V is found in the rainfall data from the latest first starting point to n hours before the first starting point. min When the rainfall data is output, the rainfall start time t s ; A5: The time after the monitoring alarm generation time t0 is recorded as t 0+1 , traverse from time t0 to t 0+1 The rainfall data at each moment is used to determine whether a rainfall greater than V can be found during this traversal. min rainfall data; yes, the last one greater than V min The collection time of rainfall data is marked as the end time of rainfall t e , with t e The time is the second starting time; otherwise, t 0+1 The time is the second starting time; A6: traverse the rainfall data from the latest second starting point to 1 hour after the second starting point again; when a rainfall greater than V can be found in this traversal process min When the rainfall data is greater than V, use the last min The collection time of rainfall data is relative to the rainfall end time t e Update and use the latest rainfall end time t e Update the second starting point time; when no value greater than V is found in this traversal process min When the rainfall data is obtained, the second starting point time is updated with the time 1 hour after the second starting point time used in this traversal; A7: Repeat step A6 until the latest rainfall ends at time t e As the starting point, no value greater than V can be found in the next n hours. min When the rainfall data is obtained; determine the rainfall end time t e Is it empty? If yes, the monitoring alarm generation time t0 is used as the rainfall end time t e Output; No, output the latest rainfall end time t e .

3. The method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk according to claim 1 is characterized in that: The calculation of the rainfall index for each rainfall includes: B1: Extract the rainfall start time t of this rainfall s and the rainfall end time t e , judge t e -t s Is it greater than 3 hours? If yes, calculate the rainfall every 3 hours using the sliding time window and compare to get t s Time t e Maximum 3-hour rainfall between and 3h ; No, jump to B2; B2: Determine t e -t s Is it greater than 2 hours and 30 minutes? If yes, use t s Time t e The accumulated rainfall between and is taken as the maximum 3-hour rainfall I 3h ; No, set the maximum 3-hour rainfall I 3h Set to null value; B3: Extract the rainfall start time t of this rainfall s and the rainfall end time t e , judge t e -t s Is it greater than 1 hour? If yes, calculate the rainfall every 1 hour using the sliding time window and compare to get t s Time t e Maximum hourly rainfall between and h ; No, jump to B4; B4: Determine t e -t s Is it greater than 40 minutes? Yes, in t s Time t e The accumulated rainfall between and is taken as the maximum hourly rainfall I h ; No, set the maximum hourly rainfall I h Set to null value.

4. The method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk according to claim 1 is characterized in that: The step of establishing a water accumulation point database comprises: Extract all rainfall data stored in the database corresponding to the current waterlogging point, and extract the rainfall period t of each rainfall data i s ~t i e , extract each rainfall period t i s ~t i e The maximum water depth h of the corresponding water point i max ; where i is the recording order of rainfall data, rainfall period t i s ~t i e The time range is from time t i s At time t i e ; The maximum water depth h i max The rainfall period is not less than 0.05 meters. i s ~t i e Mark as target time period, extract the maximum water depth h in each target time period i max The waterlogging level is located, and the critical rainfall time and critical rainfall duration of the waterlogging level are obtained by interpolation method; among them, the waterlogging level includes level IV waterlogging level: 0<h max ≤h 4,max , Grade III water accumulation level: h 4,max <h max ≤h 3,max , Level II water accumulation level: h 3,max <h max ≤h 2,max , Grade I waterlogging level: h 2,max <h max ; The maximum water depth h in the target time period corresponding to each waterlogging point is obtained. i max , water accumulation level, rainfall start time t i s , Rainfall end time t i e , critical rainfall time, critical rainfall duration, and the maximum hourly rainfall obtained according to the critical rainfall duration I h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i The water accumulation information is recorded in a water accumulation point water accumulation information statistical table to form a water accumulation point water accumulation information sample database; The maximum water depth h i max The rainfall period is less than 0.05 meters. i s ~t i e Mark as non-target time period, and calculate the maximum hourly rainfall I in each non-target time period according to steps A1 to A7 and steps B1 to B4. h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i ＜0.05 ; The maximum hourly rainfall I h , Maximum 3-hour rainfall I 3h and the cumulative rainfall V i ＜0.05 The data are recorded in the statistical table of rainfall periods without water accumulation at water accumulation points, forming a database of rainfall periods without water accumulation at water accumulation points.

5. A method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk according to claim 4, characterized in that: The method of using the interpolation method to obtain the critical rainfall moment and critical rainfall duration of the water accumulation level includes: Extract the waterlogging level and determine whether it is level IV. If yes, use the interpolation method to obtain the maximum waterlogging depth h in this record. i max The first time when the rainfall is equal to 0.05 meters is marked as the critical rainfall moment of Grade IV flooding level t i 0.05 ; No, use the interpolation method to obtain the moment when the water depth of the waterlogging point reaches the minimum value of the waterlogging level, and mark the moment as the critical rainfall moment of the waterlogging level; among them, the critical rainfall moment of level III waterlogging level is marked as t i (h4,max-Δh3,cr) , the critical rainfall moment of level II flooding is marked as t i (h3,max-Δh2,cr) , the critical rainfall moment of level I flooding is marked as t i (h2,max-Δh1,cr) ; Get the time interval of the critical rainfall moment [t j-1 ,t j ], and the time interval [t j-1 ,t j ] Total rainfall V j , calculate t using interpolation j-1 The rainfall Vz in the time period between the critical rainfall moment and the critical rainfall moment; where the time interval [t j-1 ,t j ] is the collection time of rainfall data, and z in rainfall Vz is the classification number of waterlogging level; Calculate the rainfall start time t i s Critical rainfall time to level IV flooding level t i 0.05 The critical rainfall duration Δt i 4. Calculate the rainfall start time t i s Critical rainfall time to level III flooding level t i (h4,max-Δh3,cr) The critical rainfall duration Δt i 3. Calculate the rainfall start time t i s Critical rainfall time to level II flooding level t i (h3,max-Δh2,cr) The critical rainfall duration Δt i 2. Calculate the rainfall start time t i s The critical rainfall moment for level I flooding is marked as t i (h2,max-Δh1,cr) The critical rainfall duration Δt i 1.

6. The method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk according to claim 1 is characterized in that: The critical rainfall threshold is calculated and dynamically updated to obtain a four-level threshold table of critical rainfall at waterlogging points, including: According to the latest waterlogging information sample database, the rainfall periods with the same waterlogging level are compared and the minimum maximum hourly rainfall I min h,z , minimum maximum 3-hour rainfall I min 3h,z and the minimum accumulated rainfall V min z As the initial value of the critical rainfall threshold of the corresponding level, and record the rainfall period (t i s ~t i e ) min w,z , the minimum maximum hourly rainfall I in each waterlogging level at the waterlogging point min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min And the rainfall period corresponding to each minimum value (t i s ~t i e ) min w,z Recorded in the four-level threshold table of critical rainfall at waterlogging points; among them, rainfall period (t i s ~t i e ) min w,z Where w is the classification number of rainfall, and z is the classification number of waterlogging level; The critical rainfall thresholds are dynamically updated to obtain a four-level threshold table of critical rainfall at waterlogging points.

7. A method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk according to claim 6, characterized in that: The critical rainfall threshold is dynamically updated to obtain a four-level threshold table of critical rainfall at waterlogging points, including: C1: Determine whether the IV-level threshold stored in the IV-level threshold table of critical rainfall at waterlogging points is not empty; if yes, jump to C2; if no, jump to C5; among which, non-empty means there is a specific value; C2: Extract the rainfall period corresponding to the stored IV level threshold, mark all rainfall periods after the rainfall period corresponding to the stored IV level threshold as reference rainfall periods in the latest rainfall period database without water accumulation, and count the maximum value I of the maximum hourly rainfall in the reference rainfall period. h lat_max and the second largest value I h lat_semax , the maximum value of the maximum rainfall in 3 hours I 3h lat_max and the second largest value I 3h lat_semax , the maximum value of the accumulated rainfall V lat_max and the next largest value V lat_semax ; C3: If the maximum value and the second maximum value are ≥ K times the existing IV level threshold of the corresponding indicator, that is: I h lat_max 、I h lat_semax ≥K×I min h,Ⅳ ; Some I 3h lat_max , I 3h lat_semax ≥K×I min 3h,Ⅳ ; or V lat_max 、V lat_semax ≥K×V min Ⅳ ; Among them, K is a proportional adjustment coefficient greater than 1; C4: Determine whether the maximum value and the second maximum value in the three formulas of C3 are ≥ K times the existing IV-level threshold of the corresponding indicator. If yes, mark the earliest rainfall period in the rainfall period corresponding to the established formula as the earliest rainfall period. In the waterlogging point waterlogging information sample database, if the minimum maximum hourly rainfall I with waterlogging level of IV after the earliest rainfall period can be found, min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min , the corresponding minimum value is used as the latest IV level threshold; if the minimum maximum hourly rainfall I with water accumulation level of IV after the earliest rainfall period cannot be found, min h , minimum maximum 3-hour rainfall I min 3h , minimum accumulated rainfall V min , then directly use the maximum value of the maximum hourly rainfall in the reference rainfall period I h lat_max , the maximum value of the maximum rainfall in 3 hours I 3h lat_max , the maximum value of the accumulated rainfall V lat_max As the latest level IV threshold; No, do not update the existing Level IV threshold; C5: Query the latest database of rainfall periods with or without water accumulation, and extract the maximum hourly rainfall I max h,<0.05 , Maximum 3-hour rainfall I max 3h,<0.05 and the maximum accumulated rainfall V max <0.05 , and the rainfall period corresponding to each maximum value (t i s ~t i e ) max w,<0.05 ; Determine whether the maximum value is not less than the minimum value of the threshold corresponding to each level, and the rainfall period corresponding to the maximum value is greater than the rainfall period corresponding to the minimum value of the threshold, that is: I max h,<0.05 ≥min(I min h,Ⅲ ,I min h,Ⅱ ,I min h,Ⅰ ), and (t i s ~t i e ) max h,<0.05 >min((t i s ~t i e )) min h, Ⅲ ,(t i s ~t i e ) min h,Ⅱ ,(t i s ~t i e ) min h,Ⅰ ); or I max 3h,<0.05 ≥min(I min 3h,Ⅲ , I min 3h,Ⅱ , I min 3h,Ⅰ ), and (t i s ~t i e ) max 3h,<0.05 >min((t i s ~t i e )) min 3h,Ⅲ , (t i s ~t i e )) min 3h,Ⅱ , (t i s ~t i e )) min 3h,Ⅰ ); or V max <0.05 ≥ min(V min Ⅲ , V min Ⅱ , V min Ⅰ ), and (t i s ~t i e ) max V,<0.05 > min((t i s ~t i e )) min V,Ⅲ , (t i s ~t i e )) min V,Ⅱ , (t i s ~t i e )) min V,Ⅰ ); Yes, the maximum hourly rainfall in the IV threshold is max h,Ⅳ or the maximum maximum 3-hour rainfall I max 3h,Ⅳ or the maximum accumulated rainfall V max Ⅳ Updated to correspond to the maximum hourly rainfall I max h,<0.05 or the maximum maximum 3-hour rainfall I max 3h,<0.05 or the maximum accumulated rainfall V max <0.05 , and update the corresponding rainfall period; No, no update; C6: Judge and optimize the rationality of the four-level critical rainfall threshold, and update the critical rainfall threshold of the corresponding waterlogging point after each alarm of the water depth monitoring equipment.

8. A method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk according to claim 7, characterized in that: The rationality of the four-level critical rainfall threshold is judged and optimized, including: D1: Extract the critical rainfall thresholds at each level, and determine whether each critical rainfall threshold satisfies the relationship of level IV threshold < level III threshold < level II threshold < level I threshold; if yes, do nothing; if no, that is, if the high level threshold is ≤ the low level threshold, jump to D2; D2: Determine whether the critical rainfall threshold of Level IV threshold is greater than the thresholds of the previous levels; if yes, recalculate the critical rainfall threshold of Level IV threshold, and determine the size of the critical rainfall threshold of Level IV threshold and Level III threshold, Level II threshold, and Level I threshold respectively; if there is a threshold ≤ Level IV threshold, remove the threshold at that level, and query the minimum value of that level greater than Level IV threshold in the statistical table of waterlogging information at waterlogging points to update the critical rainfall threshold of that level; if no, further determine the size relationship among Level III threshold, Level II threshold, and Level I threshold, and readjust the threshold.

9. The method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk according to claim 7, characterized in that: The water depth monitoring device at the water accumulation point updates the critical rainfall threshold of the corresponding water accumulation point after each alarm, including: E1: Extract the latest maximum hourly rainfall I obtained by this alarm from the water accumulation point water accumulation information sample database h new , Maximum 3-hour rainfall I 3h new and the cumulative rainfall V new The size of the stored threshold of the corresponding level; determine whether the stored threshold of the corresponding level is empty or greater than the latest value, that is: I h,z new <I h,z min , or I 3h,z new <I 3h,z min , or V new <V z min ; Wherein, z is the classification code of each level, z can be Ⅳ, Ⅲ, Ⅱ or Ⅰ; If yes, update the stored threshold of the corresponding level to the latest value, and record the rainfall period corresponding to the minimum value; if no, do nothing; E2: Extract each critical rainfall threshold of the updated maximum hourly rainfall, maximum 3-hour rainfall and cumulative rainfall, and determine whether each critical rainfall threshold satisfies the size distribution relationship of Level IV threshold < Level III threshold < Level II threshold < Level I threshold; if yes, do not perform any operation; if no, if the latest critical rainfall threshold is not greater than a lower level threshold of the critical rainfall threshold, then the lower level threshold of the critical rainfall threshold is eliminated; if there is a critical rainfall threshold greater than a higher level threshold of the critical rainfall threshold, then the level threshold is not updated first and is restored to the original value or null value.

10. The method for dynamic assessment of critical rainfall at a waterlogging point and prediction of waterlogging risk according to claim 1, characterized in that: The prediction of waterlogging time and risk based on the four-level threshold table of critical rainfall at waterlogging points includes: F1: Start the prediction operation and mark the start time, obtain the actual rainfall data of each waterlogging point 3 hours before the start time, and obtain the forecast rainfall data of each waterlogging point from the forecast rainfall data of the waterlogging point; among them, the actual rainfall data is obtained from the actual data of the rainfall station closest to the waterlogging point, and the forecast rainfall data of the upcoming period is the forecast rainfall data within a certain period after the start time; the forecast rainfall data is obtained from the weather forecast of the waterlogging point; F2: Calculate the rainfall in the previous hour and the previous three hours at the start time based on the actual rainfall data of the waterlogging point, and compare them with the corresponding thresholds of each level in the four-level threshold table of critical rainfall of the waterlogging point, to determine whether the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours are both less than Level IV; if yes, send out the information that there is no risk of waterlogging at the current start time, and jump to F3; if no, take the highest level of the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours as the predicted waterlogging level, take the current start time as the predicted waterlogging time, send out the information that there is a risk of waterlogging at the current start time, and output the predicted waterlogging level and predicted waterlogging time; F3: Extract the update time interval of the actual rainfall data and the forecast rainfall data; determine whether the update time interval of the actual rainfall data is less than or equal to the forecast rainfall data; if yes, mark the predictable time based on the start time and the update time interval of the forecast rainfall data; if no, jump to F4; F4: Determine whether the rainfall in the three hours before the current start time is 0; if yes, still mark the predictable time based on the start time and the update time interval of the forecast rainfall data; if no, mark the predictable time based on the start time and the update time interval of the actual rainfall data; F5: Extract predictable moments in sequence, calculate the rainfall in the previous hour and the previous three hours at the waterlogging point at the extracted predictable moment based on the actual rainfall data and forecast rainfall data of the waterlogging point, and compare them with the corresponding thresholds of each level in the four-level threshold table of critical rainfall of the waterlogging point, and judge whether the threshold levels of the 1-hour rainfall and the 3-hour rainfall are both less than Level IV; if yes, jump to F6; if no, take the highest level of the threshold levels of the rainfall in the previous hour and the rainfall in the previous three hours of the predictable moment as the predicted waterlogging level, take the current predictable moment as the predicted waterlogging moment, send out the information that there is a risk of waterlogging at the current predictable moment, and output the predicted waterlogging level and predicted waterlogging moment; F6: Extract the next predictable moment and repeat F5 step until a predictable moment with waterlogging risk is found, then issue the predicted waterlogging level and predicted waterlogging time. If no predictable moment with waterlogging risk is found, stop judging when the predictable moment extraction is completed.

Images