A flood prediction method, device and medium based on edge computing

Abstract

The embodiment of the application discloses a flood prediction method and device based on edge computing and a medium. It belongs to the technical field of data processing. The method solves the problem that data feedback to the application terminal is delayed, so that it is difficult to respond to mountain torrent disasters in time. The method comprises the following steps: acquiring historical rainstorm parameters corresponding to a current to-be-measured area; determining flood prediction results corresponding to a plurality of preset flood calculation schemes based on the plurality of preset flood calculation schemes and the historical rainstorm parameters; comparing the flood prediction results to determine a reference flood calculation scheme based on a preset screening rule; predicting the flood of the to-be-measured area based on the reference flood calculation scheme, obtaining a flood disaster grade according to a flood prediction result, and uploading the prediction result to a data center to alarm the flood prediction information of the to-be-measured area.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a flood prediction method, device and medium based on edge computing. Background Technology

[0002] Flash floods are caused by short-duration torrential rains that cause rivers to swell and flood roads, bridges, farmland, and villages, severely damaging infrastructure and the natural environment, and threatening the safety of people's property and lives.

[0003] In flash flood disaster calculation, a common problem arises where hydrological data is unavailable for design flood calculations in some areas due to factors such as small river control areas or short-term monitoring station construction. Because different regions have varying watershed characteristics and varying levels of sophistication in hydrological and rainfall monitoring systems, coupled with the inherent randomness of hydrological elements, traditional small watershed design flood calculation systems first aggregate all data before transmitting it to a cloud server in the event of heavy rainfall. In this scenario, the data processing task is concentrated in the backend, leading to a heavy processing burden on the central server and delays in data feedback to application terminals, hindering timely responses to flash flood disasters. Summary of the Invention

[0004] This application provides a flood prediction method, device, and medium based on edge computing to solve the following technical problem: In the prior art, data processing tasks are concentrated in the background, resulting in a heavy processing burden on the central server and a delay in data feedback to the application terminal, making it difficult to respond to flash flood disasters in a timely manner.

[0005] The embodiments of this application adopt the following technical solutions:

[0006] This application provides a flood prediction method based on edge computing. It includes: acquiring historical rainfall parameters corresponding to the current area to be measured; wherein the historical rainfall parameters include at least one of the following: rainfall duration, rainfall frequency, rainfall mean, rainfall variation coefficient, rainfall deviation coefficient, point-area conversion coefficient, and modulus ratio coefficient; determining the flood prediction results corresponding to each of the preset flood calculation schemes based on multiple flood calculation schemes and historical rainfall parameters; comparing the flood prediction results to determine a reference flood calculation scheme based on preset screening rules; performing flood prediction on the area to be measured based on the reference flood calculation scheme, obtaining the flood disaster level according to the flood prediction results, and uploading the prediction results to a data center to issue an alarm for the flood prediction information of the area to be measured.

[0007] This application embodiment processes data through an edge server. The edge computing model deploys the flood calculation system on an edge server close to the user, requiring only the transmission of some critical data to the cloud, reducing reliance on data transmission. Edge computing can process data streams quickly and without latency. Calculations are performed at the data edge, and the results are transmitted to the data center, saving on data communication network bandwidth costs. This reduces the amount of data transmitted over the network, thereby saving transmission costs. Furthermore, this application embodiment compares the prediction results of multiple flood calculation schemes to determine a reference flood calculation scheme, improving the accuracy of the final prediction results and enabling precise prevention of flash floods in the monitored area.

[0008] In one implementation of this application, flood prediction results corresponding to each of the pre-set flood calculation schemes are determined based on pre-set multiple flood calculation schemes and historical rainfall parameters. Specifically, this includes: determining a first flood prediction result based on a pre-set inference formula method and historical rainfall parameters; determining a second flood prediction result based on a pre-set instantaneous unit hydrograph method and historical rainfall parameters; and determining a third flood prediction result based on a pre-set hydrological model derivation method and historical rainfall parameters.

[0009] In one implementation of this application, the first flood prediction result includes at least the average rainfall intensity corresponding to different rainfall frequencies, the infiltration rate corresponding to different rainfall frequencies, the flood peak value corresponding to different rainfall frequencies, the surface flood peak value corresponding to different rainfall frequencies, the net rainfall duration corresponding to different rainfall frequencies, the runoff time corresponding to different rainfall frequencies, the total flood volume corresponding to different rainfall frequencies, and the flood peak discharge coefficient corresponding to different rainfall frequencies; the second flood prediction result includes at least the flood peak value corresponding to different rainfall frequencies, the surface flood peak value corresponding to different rainfall frequencies, the runoff time corresponding to different rainfall frequencies, the total flood volume corresponding to different rainfall frequencies, and the flood peak discharge coefficient corresponding to different rainfall frequencies; the third flood prediction result includes at least the flood peak discharge corresponding to different rainfall frequencies.

[0010] In one implementation of this application, the flood prediction results are compared to determine a reference flood calculation scheme based on preset screening rules. Specifically, this includes: sorting the peak flow in the flood prediction results based on different rainstorm frequencies; determining the flood calculation schemes corresponding to the peak flows in the middle sequence under different rainstorm frequencies; determining the flood calculation scheme to be detected based on the number of times it is in the middle sequence; adjusting the weight of the flood calculation scheme to be detected; and using the adjusted flood calculation scheme to be detected as the reference flood calculation scheme.

[0011] In one implementation of this application, the weight adjustment of the flood detection calculation scheme specifically includes: determining the prediction result corresponding to the flood detection calculation scheme; calculating the difference between each frequency characteristic value in the prediction result and the actual characteristic value of each frequency, and determining the weight corresponding to each frequency based on the difference; and adjusting the flood detection calculation scheme based on the weight.

[0012] In one implementation of this application, flood prediction is performed on the area to be measured based on a reference flood calculation scheme, and the flood disaster level is obtained based on the flood prediction results. Specifically, this includes: comparing the current flood prediction results with the hydrological information of historical flood events to determine several historical flood events based on similarity; determining the flood characteristic values ​​corresponding to several historical flood events at different time periods; comparing the current flood prediction results with the flood characteristic values ​​corresponding to several historical flood events at different time periods to determine reference historical flood events based on similarity; and determining the flood disaster level corresponding to the current flood prediction results based on the flood level corresponding to the reference historical flood events.

[0013] In one implementation of this application, the current flood forecast result is compared with the flood characteristic values ​​corresponding to several historical flood events at different time periods to determine reference historical flood events based on similarity. Specifically, this includes: dividing the current flood forecast result into multiple time periods according to chronological order, and determining the segment forecast results corresponding to each of the multiple time periods; determining the segment flood characteristic values ​​of several historical flood events within the multiple time periods; determining the differences between the segment forecast results and the multiple segment flood characteristic values ​​based on chronological order; determining the segment similarity value between the segment flood characteristic value and the segment forecast result based on the differences; and statistically analyzing the multiple segment similarity values ​​corresponding to the same historical flood event to determine the similarity between several historical flood events and the current flood forecast result based on the statistical results, thereby determining reference historical flood events based on similarity.

[0014] In one implementation of this application, the prediction results are uploaded to a data center to issue an alarm for the flood prediction information of the area to be measured. Specifically, this includes: after uploading the prediction results to the data center, analyzing the prediction results through the data center to determine the corresponding flood prediction level; and sending the flood prediction level and prediction results to the mobile terminals of the staff in the area to be measured through the data center to issue an alarm of the corresponding level to the staff.

[0015] This application provides a flood prediction device based on edge computing, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: obtain historical rainfall parameters corresponding to the current area to be measured; wherein the historical rainfall parameters include at least one of rainfall duration, rainfall frequency, rainfall mean, rainfall variation coefficient, rainfall deviation coefficient, point-area conversion coefficient, and modulus ratio coefficient; determine the flood prediction results corresponding to the preset multiple flood calculation schemes based on multiple preset flood calculation schemes and historical rainfall parameters; compare the flood prediction results to determine a reference flood calculation scheme based on preset screening rules; perform flood prediction for the area to be measured based on the reference flood calculation scheme, obtain the flood disaster level according to the flood prediction results, and upload the prediction results to the data center to issue an alarm for the flood prediction information of the area to be measured.

[0016] This application provides a non-volatile computer storage medium storing computer-executable instructions. These instructions are configured to: acquire historical rainfall parameters corresponding to the current area to be measured; wherein the historical rainfall parameters include at least one of the following: rainfall duration, rainfall frequency, rainfall mean, rainfall variation coefficient, rainfall deviation coefficient, point-area conversion coefficient, and modulus ratio coefficient; determine flood prediction results corresponding to each of the preset flood calculation schemes based on multiple flood calculation schemes and historical rainfall parameters; compare the flood prediction results to determine a reference flood calculation scheme based on preset screening rules; perform flood prediction for the area to be measured based on the reference flood calculation scheme, obtain the flood disaster level based on the flood prediction results, and upload the prediction results to a data center to issue an alarm for the flood prediction information of the area to be measured.

[0017] The above-mentioned technical solutions adopted in this application embodiment can achieve the following beneficial effects: This application embodiment processes data through an edge server. The edge computing mode deploys the flood calculation system on an edge server close to the user, requiring only the transmission of some key data to the cloud, reducing reliance on data transmission. Edge computing can process data streams quickly and without delay. Calculations are performed at the edge of the data, and the calculation results are transmitted to the data center, saving data communication network bandwidth costs. This reduces the amount of data transmitted over the network, thereby saving transmission costs. Furthermore, this application embodiment compares the prediction results of multiple flood calculation schemes to determine a reference flood calculation scheme, which can improve the accuracy of the final prediction results, thereby enabling precise prevention of flash floods in the area to be tested. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0019] Figure 1 A flowchart illustrating a flood prediction method based on edge computing, provided for an embodiment of this application;

[0020] Figure 2 This is a schematic diagram of the structure of a flood prediction device based on edge computing, provided in an embodiment of this application. Detailed Implementation

[0021] This application provides a flood prediction method, device, and medium based on edge computing.

[0022] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0023] The technical solutions proposed in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0024] Figure 1 A flowchart of a flood prediction method based on edge computing is provided for embodiments of this application, as shown below. Figure 1 As shown, the flood prediction method based on edge computing includes the following steps:

[0025] Step 101: Obtain the historical rainfall parameters corresponding to the current area to be measured; wherein, the historical rainfall parameters include at least one of the following: rainfall duration, rainfall frequency, rainfall mean, rainfall variation coefficient, rainfall deviation coefficient, point-area conversion coefficient, and modulus ratio coefficient.

[0026] In one embodiment of this application, for areas lacking long-term rainfall data, the design storm calculation is performed according to relevant storm and flood reference manuals and related atlases.

[0027] Specifically, the design rainfall duration for the watershed includes four durations: 1 hour, 3 hours, 6 hours, and 24 hours. The frequency is designed according to the specific requirements of the watershed, such as 20% (5-year return period), 10% (10-year return period), 5% (20-year return period), 2% (50-year return period), and 1% (100-year return period). The main design rainfall parameters include the mean rainfall H and the rainfall variability coefficient C. V Rainfall deviation coefficient C S Point-to-surface conversion factor α and modulus factor K P .

[0028] Step 102: Based on the preset multiple flood calculation schemes and the historical rainstorm parameters, determine the flood prediction results corresponding to the preset multiple flood calculation schemes respectively.

[0029] In one embodiment of this application, a first flood prediction result is determined based on a pre-set inference formula method and historical rainfall parameters. A second flood prediction result is determined based on a pre-set instantaneous unit hydrograph method and historical rainfall parameters. A third flood prediction result is determined based on a pre-set hydrological model derivation method and historical rainfall parameters.

[0030] In one embodiment of this application, the first flood prediction result includes at least the average rainfall intensity corresponding to different rainfall frequencies, the infiltration rate corresponding to different rainfall frequencies, the peak flood value corresponding to different rainfall frequencies, the surface peak flood value corresponding to different rainfall frequencies, the net rainfall duration corresponding to different rainfall frequencies, the runoff time corresponding to different rainfall frequencies, the total flood volume corresponding to different rainfall frequencies, and the peak flow coefficient corresponding to different rainfall frequencies. The second flood prediction result includes at least the peak flood value corresponding to different rainfall frequencies, the surface peak flood value corresponding to different rainfall frequencies, the runoff time corresponding to different rainfall frequencies, the total flood volume corresponding to different rainfall frequencies, and the peak flow coefficient corresponding to different rainfall frequencies. The third flood prediction result includes at least the peak flow rate corresponding to different rainfall frequencies.

[0031] Specifically, in the calculation of design floods using the instantaneous unit hydrograph method, the mathematical expression for the integrated instantaneous unit hydrograph method is:

[0032]

[0033] In the formula, μ(0,t) is the vertical height of the instantaneous unit line at time t; Γ(n) is the gamma function of order n; n is the watershed storage capacity, equivalent to the number of linear reservoirs; K is the storage coefficient, equivalent to the parameter of the watershed confluence time; and e is the base of the natural logarithm.

[0034] Specifically, in the derivation of design floods using hydrological models:

[0035] (1) The calculation formula of the evapotranspiration model is as follows:

[0036] E p =K·E0;

[0037] wherein, E p : Evapotranspiration capacity; E0: Measured evaporation; K: Evaporation conversion coefficient.

[0038]

[0039] wherein, C: Deep evaporation conversion coefficient; WU: Upper layer tension water storage; WL: Lower layer tension water storage; WLM: Lower layer tension water capacity; P: Precipitation; E: Calculated evaporation.

[0040] (2) Runoff yield calculation:

[0041] For runoff yield calculation, the model adopts a water storage capacity curve to consider the problem of uneven distribution of soil water deficit. The curve is a parabola with exponent b, which can be expressed by an exponential equation as follows:

[0042]

[0043] wherein, f: Runoff-producing area; F: Total watershed area; Wm: Point unsaturated zone water storage capacity; W mm : Maximum point unsaturated zone water storage capacity; b: Exponent of tension water storage capacity curve.

[0044] The average watershed water storage capacity WM can be obtained from the above equation as:

[0045]

[0046] The ordinate value a of antecedent soil moisture corresponding to W is:

[0047]

[0048] When P-E+a<Wmm, partial runoff yield occurs, then

[0049]

[0050] If the watershed contains an impermeable area, that is, when the impermeable area ratio IMP≠0, the formula for calculating WM will be changed to the following formula:

[0051]

[0052] (3) Runoff separation calculation:

[0053] The calculation formulas for runoff of each water source are as follows:

[0054] When S+R≤SM:

[0055] RS = 0

[0056] RI = (S+R)*KI*F R

[0057] RG = (S+R)*KG*F R ;

[0058] When S+R>SM:

[0059] RS = (S+R-SM)*F R

[0060] RI = SM*KI*F R

[0061] RG = SM*KG*F R ;

[0062] In the formula: S: free water; R: total runoff; SM: free water storage capacity; RS: surface runoff; RI: interflow; KI: interflow outflow coefficient; RG: groundwater runoff; KG: groundwater outflow coefficient; F R : runoff generation area ratio, that is, the bottom width of the free water storage reservoir.

[0063] Its distribution characteristic is approximated by an exponential equation, and a parabola is also adopted herein, with EX used as the exponent, whereby:

[0064]

[0065] SSM = (1+EX)SM

[0066]

[0067] When P-E+AU<SSM:

[0068]

[0069] When P-E+AU≥SSM:

[0070] RS = (P-E+S-SM)F R ;

[0071] Total runoff R can be calculated as follows:

[0072] R = RS+RI+RG;

[0073] In the formula: f: runoff generation area; F: total watershed area; S m : is the single-point free water storage capacity of the watershed; S mm: The maximum free water storage capacity at a single point in the basin; EX: The free water storage capacity-area distribution curve index of the basin; SSM: The maximum free water storage capacity at a single point in the basin.

[0074] The DEM data is used to calculate the unit line parameters of the terrain. The instantaneous unit line of the terrain is used for the runoff calculation, and the formula is as follows:

[0075]

[0076]

[0077]

[0078] Where: R A R is the area ratio. B R is the ratio of river numbers. L For the river length ratio, L Ω denoted by the superordinate river length, v is the flow velocity, Г(x) is the Gamma function, and t is time.

[0079] Step 103: Compare the flood prediction results to determine a reference flood calculation scheme based on preset screening rules.

[0080] In one embodiment of this application, the peak flow rates in flood prediction results are sorted based on different rainfall frequencies. Flood calculation schemes corresponding to peak flows in the middle sequence under different rainfall frequencies are determined. Based on the number of times a peak flow rate is in the middle sequence, a flood calculation scheme to be detected is determined, the weights of the flood calculation scheme to be detected are adjusted, and the adjusted flood calculation scheme to be detected is used as a reference flood calculation scheme.

[0081] In one embodiment of this application, the prediction result corresponding to the flood detection calculation scheme is determined. The difference between each frequency characteristic value in the prediction result and the actual characteristic value of each frequency is calculated, and the weight corresponding to each frequency is determined based on the difference. The flood detection calculation scheme is then adjusted based on the weights.

[0082] Specifically, in the embodiments of this application, the calculation results of the design flood for the area to be tested based on the reasoning formula method for design rainstorm are shown in Table 1.

[0083]

[0084] Table 1

[0085] Specifically, in the embodiments of this application, the calculation results of the design flood derived from the design storm based on the instantaneous unit hydrograph method for the area to be tested are shown in Table 2.

[0086]

[0087] Table 2

[0088] Specifically, in the embodiments of this application, the calculation results of the design flood derived from the design storm based on the Xin'anjiang model method for the area to be tested are shown in Table 3.

[0089]

[0090] Table 3

[0091] Specifically, based on different rainfall frequencies, the peak flow rates in the flood prediction results corresponding to different schemes are ranked, and the flood calculation schemes that are in the middle sequence under different rainfall frequencies are selected. Based on different rainfall frequencies, the number of times each scheme is in the middle sequence is counted, and the flood calculation scheme with the most counts is selected as the flood calculation scheme to be tested.

[0092] Furthermore, the predicted results of the obtained calculation scheme to be detected are compared with the actual feature values ​​corresponding to each rainstorm frequency. Based on the comparison results, the difference between the actual value and the value corresponding to each rainstorm frequency is determined, and then the weight of each rainstorm frequency is set according to the difference. This improves the accuracy of the prediction results.

[0093] Step 104: Based on the reference flood calculation scheme, perform flood prediction on the area to be measured, obtain the flood disaster level according to the flood prediction results, and upload the prediction results to the data center to issue an alarm for the flood prediction information of the area to be measured.

[0094] In one embodiment of this application, the current flood forecast result is compared with hydrological information of historical flood events to determine a number of historical flood events based on similarity. Flood characteristic values ​​corresponding to the several historical flood events at different time periods are determined. The current flood forecast result is compared with the flood characteristic values ​​corresponding to the several historical flood events at different time periods to determine reference historical flood events based on similarity. Based on the flood level corresponding to the reference historical flood events, the flood disaster level corresponding to the current flood forecast result is determined.

[0095] Specifically, the hydrological information of multiple historical flood events related to the area to be measured is retrieved from the database. The current flood forecast results are compared with the hydrological information of multiple historical flood events to identify several historical flood events with high similarity. These historical flood events are then divided into multiple time periods, and the flood characteristic values ​​corresponding to each time period are determined.

[0096] Specifically, based on chronological order, the current flood forecast results are divided into multiple time periods, and the corresponding segment forecast results for each time period are determined. Segment flood characteristic values ​​for several historical flood events within these multiple time periods are identified. Based on chronological order, the differences between each segment forecast result and the multiple segment flood characteristic values ​​are determined sequentially. Segment similarity values ​​between the segment flood characteristic values ​​and the segment forecast results are determined based on these differences. The segment similarity values ​​corresponding to the same historical flood event are statistically analyzed to determine the similarity between several historical flood events and the current flood forecast results, and reference historical flood events are determined based on these similarities.

[0097] For example, the current flood forecast results are divided into multiple time periods, A1, A2, A3, A4, A5, and A6, and the flood forecast characteristic values ​​for each time period are determined. The characteristic values ​​of the first historical flood event are divided into multiple time periods, B1, B2, B3, B4, B5, and B6, with each time period being the same as the time period after the current flood forecast results are divided. Similarly, the characteristic values ​​of the second historical flood event are divided into multiple time periods, C1, C2, C3, C4, C5, and C6, with each time period being the same as the time period after the current flood forecast results are divided. A1 is compared with B1, and A1 is compared with C1 to obtain the segment similarity values ​​between the forecast results and each historical result within the first time period. Similarly, the characteristic values ​​of each time period are compared sequentially to obtain the segment similarity values ​​corresponding to each historical result in different time periods. The six segment similarity values ​​corresponding to each historical result are added together to obtain the similarity score for each historical result. The historical flood event with the highest similarity score is used as the reference historical flood event.

[0098] In one embodiment of this application, after the prediction results are uploaded to the data center, the data center analyzes the prediction results to determine the corresponding flood prediction level. The data center then sends the flood prediction level and prediction results to the mobile terminals of staff in the area to be predicted, issuing an alarm of the corresponding level to the staff.

[0099] Specifically, after receiving the flood forecast results for the area to be monitored, the edge server sends the forecast results to the data center. The data center compares the received flood forecast feature values ​​with a pre-set flood forecast level table to determine the flood forecast level for the area to be monitored. Based on different forecast levels, different warnings are sent to the mobile terminals of the staff corresponding to the area to remind them to take timely precautions.

[0100] Figure 2 This is a schematic diagram of a flood prediction device based on edge computing, provided as an embodiment of this application. Figure 2 As shown, the edge computing-based flood prediction device 200 includes at least one processor 201 and a memory 202 communicatively connected to the at least one processor 201. The memory 202 stores instructions executable by the at least one processor 201, which, when executed, enable the at least one processor 201 to: acquire historical rainfall parameters corresponding to the current area to be measured; wherein the historical rainfall parameters include at least one of the following: rainfall duration, rainfall frequency, rainfall mean, rainfall variation coefficient, rainfall deviation coefficient, point-area conversion coefficient, and modulus ratio coefficient; determine the flood prediction results corresponding to each of the preset multiple flood calculation schemes and historical rainfall parameters; compare the flood prediction results to determine a reference flood calculation scheme based on preset screening rules; perform flood prediction for the area to be measured based on the reference flood calculation scheme, obtain the flood disaster level based on the flood prediction results, and upload the prediction results to a data center to issue an alarm for the flood prediction information of the area to be measured.

[0101] This application embodiment also provides a non-volatile computer storage medium storing computer-executable instructions, which are configured to: obtain historical rainfall parameters corresponding to the current area to be measured; wherein, the historical rainfall parameters include at least one of the following: rainfall duration, rainfall frequency, rainfall mean, rainfall variation coefficient, rainfall deviation coefficient, point-area conversion coefficient, and modulus ratio coefficient; determine the flood prediction results corresponding to the preset multiple flood calculation schemes based on multiple preset flood calculation schemes and historical rainfall parameters; compare the flood prediction results to determine a reference flood calculation scheme based on preset screening rules; perform flood prediction for the area to be measured based on the reference flood calculation scheme, obtain the flood disaster level according to the flood prediction results, and upload the prediction results to the data center to issue an alarm for the flood prediction information of the area to be measured.

[0102] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0103] The above descriptions are merely embodiments of this application and are not intended to limit the scope of this application. For those skilled in the art, various modifications and variations can be made to the embodiments of this application. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions in the embodiments of this application.

Claims

1. A flood prediction method based on edge computing, characterized in that, The method is executed by an edge server located in the area to be tested, and the method includes: Obtain the historical rainfall parameters corresponding to the current area to be measured; wherein, the historical rainfall parameters include at least one of the following: rainfall duration, rainfall frequency, rainfall mean, rainfall variation coefficient, rainfall deviation coefficient, point-area conversion coefficient, and modulus ratio coefficient; Based on multiple pre-set flood calculation schemes and the historical rainfall parameters, the flood prediction results corresponding to the multiple pre-set flood calculation schemes are determined respectively; The flood prediction results are compared to determine a reference flood calculation scheme based on preset screening rules; Based on the aforementioned reference flood calculation scheme, flood forecasts are performed for the area to be measured, and the flood disaster level is obtained based on the flood forecast results. The forecast results are then uploaded to the data center to issue an alarm for the flood forecast information of the area to be measured. The comparison of the flood prediction results to determine a reference flood calculation scheme based on preset screening rules specifically includes: The peak flow rates in the flood forecast results are sorted based on different rainfall frequencies; The flood calculation schemes corresponding to the peak flow in the intermediate sequence under different rainstorm frequencies were determined; Based on the number of times it appears in the middle sequence, a flood calculation scheme to be detected is determined, the flood calculation scheme to be detected is weighted and adjusted, and the adjusted flood calculation scheme to be detected is used as the reference flood calculation scheme. The weight adjustment of the flood detection calculation scheme specifically includes: The prediction results were obtained by determining the calculation scheme corresponding to the flood to be detected; The difference between each frequency feature value in the prediction result and the actual feature value of each frequency is calculated, and the weight corresponding to each frequency is determined based on the difference. The calculation scheme for the flood to be detected is adjusted based on the weights.

2. The flood prediction method based on edge computing according to claim 1, characterized in that, The process of determining the flood prediction results corresponding to each of the pre-set flood calculation schemes based on multiple pre-set flood calculation schemes and the historical rainfall parameters specifically includes: Based on the pre-set reasoning formula and the historical rainfall parameters, the prediction result of the first flood was determined; and Based on the preset instantaneous unit hydrograph method and the historical rainfall parameters, the second flood prediction result was determined; and Based on the pre-set hydrological model estimation method and the historical rainfall parameters, the prediction result of the third flood was determined.

3. The flood prediction method based on edge computing according to claim 2, characterized in that, The first flood prediction result includes at least the average rainfall intensity corresponding to different rainfall frequencies, the infiltration rate corresponding to different rainfall frequencies, the flood peak value corresponding to different rainfall frequencies, the surface flood peak value corresponding to different rainfall frequencies, the net rainfall duration corresponding to different rainfall frequencies, the runoff time corresponding to different rainfall frequencies, the total flood volume corresponding to different rainfall frequencies, and the flood peak discharge coefficient corresponding to different rainfall frequencies. The second flood forecast results include at least the flood peak value corresponding to different rainfall frequencies, the surface flood peak value corresponding to different rainfall frequencies, the confluence time corresponding to different rainfall frequencies, the total flood volume corresponding to different rainfall frequencies, and the flood peak flow coefficient corresponding to different rainfall frequencies. The third flood forecast results include at least the peak flow rates corresponding to different rainstorm frequencies.

4. The flood prediction method based on edge computing according to claim 1, characterized in that, The process of forecasting floods in the area to be measured based on the reference flood calculation scheme and obtaining the flood disaster level based on the flood forecast results specifically includes: The current flood forecast results are compared with the hydrological information of historical flood events to identify several historical flood events based on the similarity. The flood characteristic values ​​corresponding to the aforementioned historical flood events at different time periods were determined; The current flood forecast results are compared with the flood characteristic values ​​corresponding to the thousand historical flood events at different time periods to determine reference historical flood events based on similarity. Based on the flood levels corresponding to the reference historical flood events, the flood disaster level corresponding to the current flood forecast result is determined.

5. The flood prediction method based on edge computing according to claim 4, characterized in that, The step of comparing the current flood forecast result with the flood characteristic values ​​corresponding to the thousand historical flood events at different time periods to determine reference historical flood events based on similarity specifically includes: Based on the chronological order, the current flood forecast results are divided into multiple time periods, and the segment forecast results corresponding to each of the multiple time periods are determined. The characteristic values ​​of the flood sections for the aforementioned historical flood events within the various time periods were determined. Based on the chronological order, the differences between the prediction results of the respective sections and multiple flood characteristic values ​​of the respective sections are determined sequentially. Based on the difference, the segment similarity value between the segment flood characteristic value and the segment prediction result is determined; The similarity values ​​of multiple sections corresponding to the same historical flood event are statistically analyzed to determine the similarity between the several historical flood events and the current flood prediction result, and the reference historical flood event is determined based on the similarity.

6. The flood prediction method based on edge computing according to claim 1, characterized in that, Uploading the prediction results to the data center to issue an alarm for the flood prediction information of the area to be measured specifically includes: After the prediction results are uploaded to the data center, the data center sends the prediction results to the mobile terminals of the staff in the area to be tested, so as to issue an alarm of the corresponding level to the staff.

7. A flood prediction device based on edge computing, characterized in that, The device includes a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to perform the method described in any one of claims 1-6.

8. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are capable of performing the method described in any one of claims 1-6.

Images