Waterlogging risk detection method and device

Abstract

The application provides a waterlogging risk detection method and device. The method comprises the following steps: determining each risk rainfall event causing waterlogging in a to-be-detected area from each historical rainfall event corresponding to a rainfall grade; determining each target rainfall event with a waterlogging amount at a starting time less than or in a starting waterlogging amount interval from each risk rainfall event according to the predicted waterlogging amount of the future rainfall event at the starting time; determining a waterlogging risk probability of the rainfall grade in the to-be-detected area according to each target rainfall event corresponding to the rainfall grade and each historical rainfall event corresponding to the rainfall grade; and determining a target risk probability of each predicted rainfall grade of the future rainfall event of the to-be-detected area according to the waterlogging risk probability of each rainfall grade, so as to determine a waterlogging risk detection result of the to-be-detected area in the future rainfall event according to the target risk probability of each predicted rainfall grade.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of risk detection, in particular to a waterlogging risk detection method and device. BACKGROUND

[0002] In the prevention of urban waterlogging, detecting the risk area of waterlogging is an important basis for the planning and design of flood control and drainage facilities and the development of emergency plans. Currently, the detection of risk areas of waterlogging is based on the current rainfall of a certain area to determine whether the area is a risk area of waterlogging. However, this method has a serious lag, and a risk area may be determined when it has already been waterlogged, which affects the safety of urban drainage. SUMMARY

[0003] The present application aims to at least solve one of the technical problems in the related art. To this end, the present application provides a waterlogging risk detection method, which can timely warn the risk area of waterlogging and improve the safety of urban drainage.

[0004] The present application also provides a waterlogging risk detection device.

[0005] The present application also provides an electronic device.

[0006] The present application also provides a computer-readable storage medium.

[0007] The waterlogging risk detection method according to the first aspect of the present application comprises:

[0008] determining, according to the rainfall of each historical rainfall event of the to-be-detected area, each historical rainfall event corresponding to any rainfall grade;

[0009] determining, from each historical rainfall event corresponding to the rainfall grade, each risk rainfall event that causes the to-be-detected area to be waterlogged;

[0010] determining, from each risk rainfall event, each target rainfall event whose waterlogging amount at the starting time is less than or in the starting waterlogging amount interval according to the predicted waterlogging amount of the future rainfall event at the starting time;

[0011] determining, according to each target rainfall event corresponding to the rainfall grade and each historical rainfall event corresponding to the rainfall grade, the waterlogging risk probability of the rainfall grade in the to-be-detected area;

[0012] determine a target risk probability of each predicted rainfall level of the future rainfall event of the to-be-tested region according to the target risk probability of each predicted rainfall level, and determine a waterlogging risk detection result of the to-be-tested region in the future rainfall event according to the target risk probability of each predicted rainfall level;

[0013] The future rainfall event is a first future rainfall event of the to-be-tested region, and the predicted rainfall level is a predicted rainfall level of the future rainfall event.

[0014] According to the rainfall amount of each historical rainfall event of the to-be-tested region, each risk rainfall event corresponding to any rainfall level is determined, each target rainfall event with a waterlogging risk of the to-be-tested region is determined from each historical rainfall event corresponding to the rainfall level, and according to the predicted waterlogging amount of the future rainfall event at the starting time, each target rainfall event with a waterlogging risk of the to-be-tested region is determined from each risk rainfall event with a waterlogging risk of the to-be-tested region, and according to each target rainfall event corresponding to the rainfall level and each historical rainfall event corresponding to the rainfall level, the waterlogging risk probability of the rainfall level in the to-be-tested region is determined, and according to the waterlogging risk probability of each rainfall level, the target risk probability of each predicted rainfall level of the future rainfall event of the to-be-tested region is determined, and the waterlogging risk detection result of the to-be-tested region in the future rainfall event is determined according to the target risk probability of each predicted rainfall level. Therefore, before the future rainfall event occurs, the target rainfall event that can cause waterlogging can be determined according to the predicted waterlogging amount of the future rainfall event at the starting time, and the target rainfall event and the historical rainfall event of each level of the to-be-tested region are combined to predict whether the to-be-tested region has a waterlogging risk in the future rainfall event, so that the risk region with waterlogging can be warned in time, the accuracy of the prediction result is improved, and the safety of urban drainage is improved.

[0015] According to an embodiment of the present application, the waterlogging risk probability of the rainfall level in the to-be-tested region is determined according to each target rainfall event corresponding to the rainfall level and each historical rainfall event corresponding to the rainfall level, and the waterlogging risk probability of the rainfall level in the to-be-tested region is determined according to each target rainfall event corresponding to the rainfall level and each historical rainfall event corresponding to the rainfall level.

[0016] Each target rainfall event in each historical rainfall event corresponding to the rainfall level and each non-target rainfall event in each historical rainfall event are generated in time sequence to form an event sequence;

[0017] The event sequence is input into a trained long short-term memory network to obtain a prediction probability of the future event being the target rainfall event;

[0018] determine, according to the prediction probability, a waterlogging risk probability of the rainfall grade in the to-be-tested region;

[0019] The long short-term memory network is trained according to each historical event sample of any grade.

[0020] According to an embodiment of the present application, the method further comprises:

[0021] According to the waterlogging amount of the to-be-tested region at the current time, the drainage rate of the to-be-tested region, and the starting time of the future rainfall event, the prediction waterlogging amount is determined.

[0022] According to an embodiment of the present application, the target risk probability of the to-be-tested region in each prediction rainfall grade of the future rainfall event is determined according to the waterlogging risk probability of each rainfall grade, comprising:

[0023] The initial risk probability of the to-be-tested region in each prediction rainfall grade of the future rainfall event is determined according to the waterlogging risk probability of each rainfall grade.

[0024] The target risk probability of the prediction rainfall grade is obtained according to the prediction probability of the prediction rainfall grade and the initial risk probability of the prediction rainfall grade.

[0025] The prediction probability is a probability that a prediction rainfall amount of the future rainfall event reaches the prediction rainfall grade.

[0026] According to an embodiment of the present application, the waterlogging risk detection result of the to-be-tested region in the future rainfall event is determined according to the target risk probability of each prediction rainfall grade, comprising:

[0027] The target risk probability of each prediction rainfall grade is superimposed to obtain a waterlogging probability of the to-be-tested region in the future rainfall event.

[0028] The waterlogging risk detection result of the to-be-tested region in the future rainfall event is determined according to the waterlogging probability.

[0029] According to an embodiment of the present application, the method further comprises:

[0030] The waterlogging risk detection result is determined as the to-be-tested region having a waterlogging risk in the future rainfall event, and prompt information prompting to reconstruct a drainage pipe network of the to-be-tested region is generated.

[0031] According to an embodiment of the present application, the rainfall amount and the rainfall grade are positively correlated.

[0032] The waterlogging risk detection device according to the second embodiment of the present application comprises:

[0033] The rainfall event acquisition module is configured to determine each historical rainfall event corresponding to any rainfall grade according to rainfall amounts of each historical rainfall event of the to-be-tested region;

[0034] The first event determination module is configured to determine each risk rainfall event causing waterlogging in the to-be-tested region from each historical rainfall event corresponding to the rainfall grade;

[0035] The second event determination module is configured to determine each target rainfall event with a waterlogging amount at the starting time less than or equal to a predicted waterlogging amount of a future rainfall event at the starting time from each risk rainfall event according to the predicted waterlogging amount of the future rainfall event at the starting time;

[0036] The risk probability determination module is configured to determine a waterlogging risk probability of the rainfall grade in the to-be-tested region according to each target rainfall event corresponding to the rainfall grade and each historical rainfall event corresponding to the rainfall grade;

[0037] The waterlogging risk detection module is configured to determine a target risk probability of each predicted rainfall grade of a future rainfall event of the to-be-tested region according to the waterlogging risk probability of each rainfall grade, and to determine a waterlogging risk detection result of the to-be-tested region in the future rainfall event according to the target risk probability of each predicted rainfall grade;

[0038] The future rainfall event is a first rainfall event of the to-be-tested region in the future, and the predicted rainfall grade is a predicted rainfall grade of the future rainfall event.

[0039] The electronic device according to the third aspect of the present application comprises a processor and a memory storing a computer program, and the processor implements the waterlogging risk detection method of any one of the above embodiments when executing the computer program.

[0040] The computer readable storage medium according to the fourth aspect of the present application stores a computer program, and the computer program is executed by a processor to implement the waterlogging risk detection method of any one of the above embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0042] Figure 1 The first flowchart of the waterlogging risk detection method of some embodiments of the present application;

[0043] Figure 2 A second flowchart of a waterlogging risk detection method for some embodiments of the present application;

[0044] Figure 3 A structural schematic diagram of a waterlogging risk detection device provided by some embodiments of the present application;

[0045] Figure 4 A structural schematic diagram of an electronic device provided by some embodiments of the present application. DETAILED DESCRIPTION

[0046] In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described in detail below with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0047] In the following, the waterlogging risk detection method and device provided by the embodiments of the present application will be described and explained in detail through several specific embodiments.

[0048] In an embodiment, a waterlogging risk detection method is provided, which is applied to a terminal device and used for predicting a risk of waterlogging in a to-be-detected area. The terminal device can be any terminal device such as a mobile terminal, a desktop terminal or a server. The server can be an independent server or a server cluster composed of multiple servers, or a cloud server providing basic cloud computing services such as cloud service, cloud message database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and large message data and artificial intelligence sampling point device.

[0049] As shown in Figure 1 The waterlogging risk detection method provided by the present embodiment includes:

[0050] According to the rainfall of each historical rainfall event of the to-be-detected area, each historical rainfall event corresponding to any rainfall grade is determined.

[0051] From each historical rainfall event corresponding to the rainfall grade, each risk rainfall event causing waterlogging in the to-be-detected area is determined.

[0052] According to the predicted water accumulation amount of the future rainfall event at the starting time, each target rainfall event with a water accumulation amount less than or equal to the predicted water accumulation amount at the starting time is determined from each risk rainfall event.

[0053] determining, according to each of the target rainfall events corresponding to the rainfall grade and each of the historical rainfall events corresponding to the rainfall grade, a waterlogging risk probability of the rainfall grade in the to-be-tested region;

[0054] determining, according to the waterlogging risk probability of each of the rainfall grades, a target risk probability of each predicted rainfall grade of a future rainfall event of the to-be-tested region, so as to determine a waterlogging risk detection result of the to-be-tested region in the future rainfall event according to the target risk probability of each predicted rainfall grade;

[0055] wherein the future rainfall event is a first future rainfall event of the to-be-tested region, and the predicted rainfall grade is a predicted rainfall grade of the future rainfall event.

[0056] In some embodiments, the to-be-tested region can be any region of a city. For example, a city map can be gridded, and one of the grids after gridding can be determined as the to-be-tested region.

[0057] For a to-be-tested region, the rainfall grade of each historical rainfall event can be determined according to the rainfall amount interval in which the rainfall amount of each historical rainfall event is located. Wherein, a rainfall event refers to one rainfall, and the rainfall event can include the starting time of the rainfall, the ending time of the rainfall, and the rainfall amount of the rainfall. The rainfall amount interval is positively correlated with the rainfall grade, for example, the rainfall amount intervals are light rain (24-hour rainfall total amount less than 10 mm), moderate rain (24-hour rainfall total amount 10-25 mm), heavy rain (24-hour rainfall total amount 25-50 mm), rainstorm (24-hour rainfall total amount 50-100 mm), heavy rainstorm (24-hour rainfall total amount 100-250 mm), and extremely heavy rainstorm (24-hour rainfall total amount greater than or equal to 250 mm), and the corresponding rainfall grades are 1-6.

[0058] After obtaining the rainfall grade of the historical rainfall event, each historical rainfall event corresponding to any rainfall grade can be determined. After obtaining each historical rainfall event corresponding to a rainfall grade k, the waterlogging amount of each historical rainfall event at the starting time, i.e., the time when the rainfall event is triggered, is divided into a corresponding starting waterlogging amount interval r i wherein i represents the i-th starting waterlogging amount interval r.

[0059] The accumulated water amount of each historical rainfall event s1-s5 at the starting time can be obtained. For example, if the rainfall time of the historical rainfall event s1 is xx year xx month xx day 8:00-9:00, then xx year xx month xx day 8:00 is the starting time of the historical rainfall event s1, and at this time, the accumulated water amount of the to-be-tested region at xx year xx month xx day 8:00 can be obtained as the accumulated water amount of the historical rainfall event s1 at the starting time.

[0060] After obtaining the accumulated water amount of each historical rainfall event at the starting time, the accumulated water amount of each historical rainfall event at the starting time can be divided into the corresponding starting accumulated water amount interval according to the accumulated water amount of each historical rainfall event at the starting time. Each starting accumulated water amount interval can be set according to actual conditions, for example, each starting accumulated water amount interval r i From low to high, they can be <10 cm, 10-20 cm, 20-30 cm, and >30 cm.

[0061] After obtaining the starting accumulated water amount interval in which the accumulated water amount of each historical rainfall event at the starting time is located, the number of historical rainfall events corresponding to each starting accumulated water amount interval can be determined, so as to determine the accumulated water occurrence probability of each starting accumulated water amount interval r i of the to-be-tested region at the starting time according to the number of historical rainfall events corresponding to each starting accumulated water amount interval. i From low to high, they can be r1: <10 cm, r2: 10-20 cm, r3: 20-30 cm, and r4: >30 cm. The total number of historical rainfall events corresponding to the rainfall level k is C, and the number of historical rainfall events corresponding to each starting accumulated water amount interval r i is C1, C2, C3, and C4, respectively, and C1+C2+C3+C4=C. The accumulated water occurrence probability of each starting accumulated water amount interval r i of the to-be-tested region at the starting time is C1 / C, C1+C2 / C, C1+C2+C3 / C, and C1+C2+C3+C4 / C, respectively.

[0062] For future rainfall events, the predicted accumulated water amount of the future rainfall event at the starting time can be predicted through weather forecasting.

[0063] To further improve the accuracy of the prediction result, in some embodiments, the predicted accumulated water amount can be determined according to the accumulated water amount of the to-be-tested region at the current time, the drainage rate of the to-be-tested region, and the starting time of the future rainfall event.

[0064] For example, the accumulated water amount R t1and the drainage rate v of the to-be-tested region, and according to the weather forecast, the starting time t2 of the next rainfall event in the to-be-tested region is predicted, so as to obtain the predicted waterlogging amount of the to-be-tested region at the starting time of the future rainfall event according to the current time t1, the waterlogging amount R of the current time t1, the drainage rate v of the to-be-tested region, and the starting time t2 t1 , the drainage rate v of the to-be-tested region, and the starting time t2, the predicted waterlogging amount of the to-be-tested region at the starting time of the future rainfall event is obtained as follows:

[0065] R' = R t1 -v × (t2-t1).

[0066] After obtaining the predicted waterlogging amount at the starting time of the future rainfall event, all the risk rainfall events in which the waterlogging amount at the starting time is less than or in the starting waterlogging amount interval can be determined as target rainfall events from the risk rainfall events according to the starting waterlogging amount interval in which the predicted waterlogging amount is located, so as to determine the waterlogging risk probability corresponding to the rainfall grade in the to-be-tested region according to the number of the target rainfall events and the number of the historical rainfall events corresponding to the rainfall grade.

[0067] For example, it is assumed that the starting waterlogging amount interval in which the predicted waterlogging amount is located is r3, the total number of the historical rainfall events corresponding to the rainfall grade k is C, and the number of the historical rainfall events corresponding to each starting waterlogging amount interval r i is respectively r1: C1, r2: C2, r3: C3, and r4: C4. Since r1 and r2 are less than r3, the number of the target rainfall events is C1+C2+C3 at this time, and the waterlogging occurrence probability corresponding to the starting waterlogging amount interval r3 is (C1+C2+C3) / C. The waterlogging occurrence probability can be determined as the waterlogging risk probability corresponding to the rainfall grade k in the to-be-tested region. In this way, the waterlogging risk probability corresponding to each rainfall grade can be obtained.

[0068] In some embodiments, all the possible predicted rainfall amounts of the future rainfall event, i.e., the future nearest rainfall event, of the to-be-tested region are determined through the weather forecast, and each predicted rainfall grade of the future rainfall event is determined from the correspondence between the rainfall amount interval and the rainfall grade according to the rainfall amount interval in which each predicted rainfall amount is located.

[0069] For example, assuming that the rainfall intervals are light rain (total rainfall in 24 hours less than 10 mm), moderate rain (total rainfall in 24 hours 10-25 mm), heavy rain (total rainfall in 24 hours 25-50 mm), rainstorm (total rainfall in 24 hours 50-100 mm), heavy rainstorm (total rainfall in 24 hours 100-250 mm), and extremely heavy rainstorm (total rainfall in 24 hours greater than or equal to 250 mm), the corresponding rainfall grades are 1-6, and through weather forecasting, it is determined that the corresponding rainfall intervals of each predicted rainfall amount of the future rainfall event are moderate rain and heavy rain, i.e., the future rainfall event can be moderate rain or heavy rain, and then it can be determined that the predicted rainfall grades of the future rainfall event are 2 and 3, respectively.

[0070] After obtaining each predicted rainfall grade of the future rainfall event, the target risk probability of each predicted rainfall grade can be determined according to the corresponding relationship between the rainfall grade and the waterlogging risk probability. For example, assuming that the corresponding relationship between the rainfall grade and the waterlogging risk probability is: the waterlogging risk probability corresponding to the rainfall grade 1 is P1; the waterlogging risk probability corresponding to the rainfall grade 2 is P2; the waterlogging risk probability corresponding to the rainfall grade 3 is P3; the waterlogging risk probability corresponding to the rainfall grade 4 is P4; the waterlogging risk probability corresponding to the rainfall grade 5 is P5; and the waterlogging risk probability corresponding to the rainfall grade 6 is P6. Assuming that the predicted rainfall grades of the future rainfall event are 2 and 3, the target risk probabilities of the predicted rainfall grades are P2 and P3, respectively.

[0071] After obtaining the target risk probability of each predicted rainfall grade, if the target risk probability of one of the predicted rainfall grades is greater than a preset probability, such as greater than 50%, it indicates that the to-be-tested region is likely to have waterlogging in the future rainfall event, and at this time, it can be determined that the waterlogging risk detection result is that there is a waterlogging risk, and the to-be-tested region is a risk region in the future rainfall event; otherwise, it can be determined that the waterlogging risk detection result is that there is no waterlogging risk, and the to-be-tested region is a non-risk region in the future rainfall event.

[0072] By determining the historical rainfall events corresponding to any rainfall level based on the rainfall of each historical rainfall event in the area to be tested, and determining the risk rainfall events that may cause waterlogging in the area to be tested from the historical rainfall events corresponding to the rainfall level, and determining the target rainfall events whose waterlogging at the starting moment is less than or within the starting waterlogging interval according to the predicted waterlogging at the starting moment of the future rainfall event from the risk rainfall events, the waterlogging risk probability of the rainfall level in the area to be tested is determined based on the target rainfall events corresponding to the rainfall level and the historical rainfall events corresponding to the rainfall level, and based on the waterlogging risk probability of each rainfall level, determine the target risk probability of each predicted rainfall level of future rainfall events in the area to be tested, and based on the target risk probability of each predicted rainfall level, determine the waterlogging risk detection result of the area to be tested in the future rainfall event according to the target risk probability of each predicted rainfall level. Therefore, before a future rainfall event occurs, the target rainfall event that will cause urban flooding can be determined by the predicted water accumulation at the starting time of the future rainfall event. By combining the target rainfall events of various levels and historical rainfall events in the test area, it is possible to predict whether the test area will be at risk of urban flooding during future rainfall events. This will enable timely early warning of areas at risk of urban flooding, improve the accuracy of the prediction results, and thereby improve the safety of urban drainage.

[0073] To further improve the accuracy of the detection results, in some embodiments, Figure 2 As shown, according to each target rainfall event corresponding to the rainfall level and each historical rainfall event corresponding to the rainfall level, determining the flooding risk probability of the rainfall level in the test area includes:

[0074] Step 201: Generate an event sequence by chronologically arranging the target rainfall events in the historical rainfall events corresponding to the rainfall levels and the non-target rainfall events in the historical rainfall events.

[0075] Step 202: input the event sequence into a trained long short-term memory network to obtain a predicted probability that the future event will be the target rainfall event;

[0076] Step 203: determining the waterlogging risk probability of the rainfall level in the test area based on the predicted probability;

[0077] The long short-term memory network is trained based on historical event samples of any level.

[0078] In some embodiments, for the training of the long short-term memory network, any level of each historical rainfall event can be obtained as each historical event sample, and each historical event sample is labeled, such as being labeled as a risk sample for causing waterlogging in the to-be-detected region, and a non-risk sample for not causing waterlogging in the to-be-detected region.

[0079] Then, the labeled historical event samples are sorted in chronological order from early to late to obtain a historical event sample sequence. After obtaining the historical event sample sequence, each historical event sample in the historical event sample sequence is input into the long short-term memory network to be trained for multiple times. In each training, the first n historical event samples in the historical event sample sequence are input into the long short-term memory network, and the prediction probability of the (n+1)th historical event sample output by the long short-term memory network is obtained. Then, the prediction probability is matched with the actual probability of the (n+1)th historical event sample being a risk sample, the loss of the long short-term memory network is calculated and back propagated, and the network parameters of the long short-term memory network are adjusted. After that, the first n+1 historical event samples in the historical event sample sequence are input into the long short-term memory network for the next training, and so on. Until the prediction probability obtained by inputting the first n historical event samples in the historical event sample sequence into the long short-term memory network each time is matched with the actual probability of the (n+1)th historical event sample being a risk sample, the training of the long short-term memory network is completed.

[0080] After obtaining the trained long short-term memory network, each target rainfall event in each historical rainfall event corresponding to a rainfall level and each non-target rainfall event in each historical rainfall event can be sorted in chronological order from early to late to generate an event sequence. The non-target rainfall event is an event in each historical rainfall event except the target rainfall event. Then, the event sequence is input into the trained long short-term memory network to obtain the prediction probability of a future event being a target rainfall event, so as to use the prediction probability as the waterlogging risk probability of the rainfall level in the to-be-detected region. Thus, the waterlogging risk probability of the rainfall level can be accurately predicted through the change trend of each historical rainfall event corresponding to the rainfall level, and the accuracy of the subsequent determination of the waterlogging risk detection result of the to-be-detected region in a future rainfall event using the waterlogging risk probability of the rainfall level is improved.

[0081] Considering that the occurrence probability of each predicted rainfall level is not the same in a future rainfall event, in order to further improve the accuracy of the detection result, in some embodiments, the target risk probability of each predicted rainfall level of the to-be-detected region in a future rainfall event is determined according to the waterlogging risk probability of each rainfall level, which includes:

[0082] determining initial risk probabilities of each predicted rainfall level of the future rainfall event in the to-be-tested region according to the waterlogging risk probabilities of each rainfall level;

[0083] obtaining target risk probabilities of the predicted rainfall levels according to predicted probabilities of the predicted rainfall levels and the initial risk probabilities of the predicted rainfall levels;

[0084] The predicted probability is a probability that a predicted rainfall amount of the future rainfall event reaches the predicted rainfall level.

[0085] In some embodiments, probabilities Pi that a predicted rainfall amount of the future rainfall event in the to-be-tested region is in different rainfall amount intervals can be determined according to a weather forecast. After the probabilities Pi that the predicted rainfall amount of the future rainfall event is in different rainfall amount intervals are obtained, predicted probabilities Pi that the future rainfall event in the to-be-tested region is in different predicted rainfall levels can be determined according to a correspondence between the rainfall amount intervals and the predicted rainfall levels. For example, a probability Pi of a rainfall amount interval is determined as a predicted probability Pi of a predicted rainfall level corresponding to the rainfall amount interval.

[0086] For example, assuming that according to a weather forecast, a probability that a predicted rainfall amount of the future rainfall event in the to-be-tested region is light rain (less than 10 mm of total rainfall in 24 hours) is 10%, a probability that the predicted rainfall amount is moderate rain (10-25 mm of total rainfall in 24 hours) is 40%, and a probability that the predicted rainfall amount is heavy rain (25-50 mm of total rainfall in 24 hours) is 40%, it can be determined that a predicted probability that the predicted rainfall level of the future rainfall event is level 1 is 10%, a predicted probability that the predicted rainfall level is level 2 is 40%, and a predicted probability that the predicted rainfall level is level 3 is 40%.

[0087] Meanwhile, after the waterlogging risk probabilities of each rainfall level are obtained, waterlogging risk probabilities corresponding to each predicted rainfall level of the future rainfall event can be obtained as initial risk probabilities Pi’ of each predicted rainfall level according to a correspondence between each rainfall level and each waterlogging risk probability. For example, each rainfall level is 1-6, and corresponding waterlogging risk probabilities are P1’-P6’. If each predicted rainfall level of the future rainfall event is 1-3, the waterlogging risk probability P1’ is determined as the initial risk probability of the predicted rainfall level 1, the waterlogging risk probability P2’ is determined as the initial risk probability of the predicted rainfall level 2, and the waterlogging risk probability P3’ is determined as the initial risk probability of the predicted rainfall level 3.

[0088] After the initial risk probabilities Pi’ and the predicted probabilities Pi of each predicted rainfall level are obtained, for any predicted rainfall level k, a target risk probability P(k) thereof can be obtained according to the initial risk probability Pi’ and the predicted probability Pi thereof, that is, P(k) = Pi’ × Pi.

[0089] The initial risk probability of the to-be-tested region in each predicted rainfall level of the future rainfall event is determined according to the waterlogging risk probability of each rainfall level, the target risk probability of each predicted rainfall level is obtained according to the predicted probability of the predicted rainfall level and the initial risk probability of the predicted rainfall level, and thus the target risk probability of the predicted rainfall level obtained is combined with the occurrence probability of the predicted rainfall level, the target risk probability of the predicted rainfall level obtained is more accurate, and the accuracy of the waterlogging risk detection result obtained by using the target risk probability of the predicted rainfall level is further improved.

[0090] After the target risk probability of each predicted rainfall level is obtained, it can be detected whether there is a target risk probability greater than a preset probability, such as greater than 50%, in the target risk probability of each predicted rainfall level. If yes, it is determined that the waterlogging risk detection result is that the to-be-tested region has a waterlogging risk in the future rainfall event, and the to-be-tested region is a risk region in the future rainfall event. Otherwise, it is determined that the waterlogging risk detection result is that the to-be-tested region has no waterlogging risk in the future rainfall event.

[0091] To further improve the accuracy of the waterlogging risk detection result, in some embodiments, the waterlogging risk detection result of the to-be-tested region in the future rainfall event is determined according to the target risk probability of each predicted rainfall level, including:

[0092] The target risk probability of each predicted rainfall level is superimposed to obtain the waterlogging probability of the to-be-tested region in the future rainfall event;

[0093] The waterlogging risk detection result of the to-be-tested region in the future rainfall event is determined according to the waterlogging probability.

[0094] In some embodiments, after the target risk probability P(k) of each predicted rainfall level is obtained, the target risk probability P(k) of each predicted rainfall level is superimposed, that is, the target risk probability P(k) of each predicted rainfall level is added to obtain the waterlogging probability of the to-be-tested region in the future rainfall event. For example, assuming that the predicted rainfall levels are 1-3, the waterlogging probability of the to-be-tested region in the future rainfall event is

[0095] After the waterlogging probability of the to-be-tested region in the future rainfall event is obtained, it can be detected whether the waterlogging probability is greater than a preset probability, such as 50%. If yes, it is determined that the to-be-tested region is prone to waterlogging in the future rainfall event, and the waterlogging risk detection result is that the to-be-tested region has a waterlogging risk in the future rainfall event, and the to-be-tested region is a risk region, and an alarm information is generated.

[0096] The waterlogging probability of the to-be-tested region in a future rainfall event is obtained by superimposing the target risk probabilities of each predicted rainfall grade, so as to determine the waterlogging risk detection result of the to-be-tested region in the future rainfall event according to the waterlogging probability, thereby making the waterlogging risk detection of the to-be-tested region consider the risks of the to-be-tested region under different predicted rainfall grades, and further improving the accuracy of the waterlogging risk detection of the to-be-tested region.

[0097] To further improve the safety of drainage, in some embodiments, the method further comprises:

[0098] determining that the waterlogging risk detection result is that the to-be-tested region has a waterlogging risk in the future rainfall event, and generating prompt information prompting to reform the drainage pipe network of the to-be-tested region.

[0099] In some embodiments, the waterlogging risk detection result of the to-be-tested region in the future rainfall event is that the to-be-tested region has a waterlogging risk in the future rainfall event, and at this time, the pipe model of the drainage pipe network located in the to-be-tested region can be obtained according to the location information of the to-be-tested region, and prompt information prompting to reform the drainage pipe network of the to-be-tested region is generated, and the pipe model of the drainage pipe network is marked to prompt fine reform of the drainage pipe network.

[0100] The waterlogging risk detection device provided in the present application is described below, and the waterlogging risk detection device described below can be referred to in correspondence with the waterlogging risk detection method described above.

[0101] In an embodiment, as shown in Figure 3 a waterlogging risk detection device is provided, comprising:

[0102] The rainfall event acquisition module 210 is configured to determine, according to the rainfall amounts of each historical rainfall event of a to-be-tested region, each historical rainfall event corresponding to any rainfall grade.

[0103] The first event determination module 220 is configured to determine, from each historical rainfall event corresponding to the rainfall grade, each risk rainfall event causing the to-be-tested region to appear waterlogging.

[0104] The second event determination module 230 is configured to determine, from each risk rainfall event, each target rainfall event having a waterlogging amount at a starting time less than or equal to a predicted waterlogging amount of a future rainfall event at the starting time.

[0105] The risk probability determination module 240 is configured to determine, according to each target rainfall event corresponding to the rainfall grade and each historical rainfall event corresponding to the rainfall grade, a waterlogging risk probability of the rainfall grade in the to-be-tested region.

[0106] The waterlogging risk detection module 250 is configured to determine a target risk probability of each predicted rainfall level of the future rainfall event of the to-be-detected region according to the waterlogging risk probability of each rainfall level, and determine a waterlogging risk detection result of the to-be-detected region in the future rainfall event according to the target risk probability of each predicted rainfall level.

[0107] The future rainfall event is a first future rainfall event of the to-be-detected region, and the predicted rainfall level is a predicted rainfall level of the future rainfall event.

[0108] According to the rainfall amount of each historical rainfall event of the to-be-detected region, each historical rainfall event corresponding to any rainfall level is determined, each risk rainfall event causing waterlogging of the to-be-detected region is determined from each historical rainfall event corresponding to the rainfall level, and after determining each target rainfall event with a starting waterlogging amount less than or in a starting waterlogging amount interval of a predicted waterlogging amount of the future rainfall event at a starting time, the waterlogging risk probability of the rainfall level in the to-be-detected region is determined according to each target rainfall event corresponding to the rainfall level and each historical rainfall event corresponding to the rainfall level, and the target risk probability of each predicted rainfall level of the future rainfall event of the to-be-detected region is determined according to the waterlogging risk probability of each rainfall level, and the waterlogging risk detection result of the to-be-detected region in the future rainfall event is determined according to the target risk probability of each predicted rainfall level. Thus, before the future rainfall event occurs, the target rainfall event causing waterlogging can be determined according to the predicted waterlogging amount of the future rainfall event at the starting time, the target rainfall event and the historical rainfall event of each level of the to-be-detected region are combined to predict whether the to-be-detected region has a waterlogging risk in the future rainfall event, so that the risk region with waterlogging can be warned in time, the accuracy of the prediction result is improved, and the safety of urban drainage is improved.

[0109] In an embodiment, the risk probability determination module 240 is specifically configured to:

[0110] generate an event sequence according to each target rainfall event in each historical rainfall event corresponding to the rainfall level and each non-target rainfall event in each historical rainfall event in time sequence;

[0111] input the event sequence into the trained long short-term memory network to obtain a prediction probability of the future event being the target rainfall event;

[0112] determine the waterlogging risk probability of the rainfall level in the to-be-detected region according to the prediction probability;

[0113] The long short-term memory network is trained according to each historical event sample of any level.

[0114] In an embodiment, the second event determination module 230 is further configured to:

[0115] According to the waterlogging amount of the to-be-tested region at the current time, the drainage rate of the to-be-tested region, and the starting time of the future rainfall event, the predicted waterlogging amount is determined.

[0116] In an embodiment, the waterlogging risk detection module 250 is specifically configured to:

[0117] According to the waterlogging risk probability of each rainfall level, an initial risk probability of the to-be-tested region at each predicted rainfall level of the future rainfall event is determined.

[0118] According to the predicted probability of the predicted rainfall level and the initial risk probability of the predicted rainfall level, a target risk probability of the predicted rainfall level is obtained.

[0119] The predicted probability is a probability that a predicted rainfall amount of the future rainfall event reaches the predicted rainfall level.

[0120] In an embodiment, the waterlogging risk detection module 250 is specifically configured to:

[0121] The target risk probabilities of each predicted rainfall level are superimposed to obtain a waterlogging probability of the to-be-tested region in the future rainfall event.

[0122] According to the waterlogging probability, a waterlogging risk detection result of the to-be-tested region in the future rainfall event is determined.

[0123] In an embodiment, the waterlogging risk detection module 250 is further configured to:

[0124] The waterlogging risk detection result is determined to be that the to-be-tested region has a waterlogging risk in the future rainfall event, and prompt information prompting to reconstruct a drainage pipe network of the to-be-tested region is generated.

[0125] In an embodiment, the rainfall amount and the rainfall level are positively correlated.

[0126] Figure 4 An example of an entity structure diagram of an electronic device is shown in FIG. 1. Figure 4As shown, the electronic device can include a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 complete mutual communication through the communication bus 840. The processor 810 can call the computer program in the memory 830 to execute the waterlogging risk detection method, for example, including:

[0127] According to the rainfall of each historical rainfall event of the to-be-tested region, determine the historical rainfall events corresponding to any rainfall grade;

[0128] From each of the historical rainfall events corresponding to the rainfall grade, determine each risk rainfall event that causes the to-be-tested region to appear waterlogging;

[0129] According to the starting water accumulation interval in which the predicted water accumulation amount of the future rainfall event at the starting time is located, from each of the risk rainfall events, determine each target rainfall event whose water accumulation amount at the starting time is less than or in the starting water accumulation interval;

[0130] According to each of the target rainfall events corresponding to the rainfall grade, and each of the historical rainfall events corresponding to the rainfall grade, determine the waterlogging risk probability of the rainfall grade in the to-be-tested region;

[0131] According to the waterlogging risk probability of each rainfall grade, determine the target risk probability of each predicted rainfall grade of the future rainfall event of the to-be-tested region, so as to determine the waterlogging risk detection result of the to-be-tested region in the future rainfall event according to the target risk probability of each predicted rainfall grade;

[0132] Wherein, the future rainfall event is the first rainfall event of the to-be-tested region in the future, and the predicted rainfall grade is the predicted rainfall grade of the future rainfall event.

[0133] Further, the logic instructions in the memory 830 described above can be implemented in the form of software functional units and sold or used as independent products, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that make contributions to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the method of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0134] In another aspect, the embodiments of the present application also provide a storage medium, the storage medium includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program is executed by a processor, and the computer can execute the flood risk detection method provided by the above-mentioned embodiments, for example, including:

[0135] According to the rainfall of each historical rainfall event of the to-be-tested region, determine each historical rainfall event corresponding to any rainfall grade;

[0136] From each historical rainfall event corresponding to the rainfall grade, determine each risk rainfall event that causes the to-be-tested region to appear flooding;

[0137] According to the starting water accumulation interval in which the predicted water accumulation amount of the future rainfall event at the starting time is located, from each risk rainfall event, determine each target rainfall event whose water accumulation amount at the starting time is less than or in the starting water accumulation interval;

[0138] According to each target rainfall event corresponding to the rainfall grade, and each historical rainfall event corresponding to the rainfall grade, determine the flooding risk probability of the rainfall grade in the to-be-tested region;

[0139] According to the flooding risk probability of each rainfall grade, determine the target risk probability of each predicted rainfall grade of the future rainfall event of the to-be-tested region, so as to determine the flood risk detection result of the to-be-tested region in the future rainfall event according to the target risk probability of each predicted rainfall grade;

[0140] Wherein, the future rainfall event is the first rainfall event of the to-be-tested region in the future, and the predicted rainfall grade is the predicted rainfall grade of the future rainfall event.

[0141] The device embodiments described above are merely illustrative, wherein the units illustrated as separate components can or can not be physically separate, and the components illustrated as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0142] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, and the computer software products can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and include a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0143] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting waterlogging risk, characterized in that: include: Determining each historical rainfall event corresponding to any rainfall level based on the rainfall of each historical rainfall event in the test area; Determining, from the historical rainfall events corresponding to the rainfall levels, risk rainfall events that may cause waterlogging in the tested area; According to the starting water volume range of the predicted water volume of the future rainfall event at the starting time, determine, from the risk rainfall events, target rainfall events whose water volume at the starting time is less than or within the starting water volume range; Determining, in the test area, a waterlogging risk probability of the rainfall level according to each target rainfall event corresponding to the rainfall level and each historical rainfall event corresponding to the rainfall level; Determining target risk probabilities for each predicted rainfall level of future rainfall events in the test area based on the waterlogging risk probabilities for each rainfall level, and determining waterlogging risk detection results for the test area in the future rainfall events based on the target risk probabilities for each predicted rainfall level; The future rainfall event is the first rainfall event in the measured area in the future, and the predicted rainfall level is the predicted rainfall level of the future rainfall event.

2. The method for detecting waterlogging risk according to claim 1, wherein: Determining, according to each target rainfall event corresponding to the rainfall level and each historical rainfall event corresponding to the rainfall level, a flooding risk probability of the rainfall level in the test area, including: Generating an event sequence by chronologically sequencing the target rainfall events in the historical rainfall events corresponding to the rainfall levels and the non-target rainfall events in the historical rainfall events; Inputting the event sequence into a trained long short-term memory network to obtain a predicted probability that the future event is the target rainfall event; Determining, in the test area, a waterlogging risk probability of the rainfall level based on the predicted probability that the future event is a target rainfall event; The long short-term memory network is trained based on historical event samples of any level.

3. The method for detecting waterlogging risk according to claim 1, wherein: Also includes: The predicted amount of accumulated water is determined based on the amount of accumulated water in the area to be measured at the current moment, the drainage rate of the area to be measured, and the starting moment of the future rainfall event.

4. The method for detecting waterlogging risk according to any one of claims 1 to 3, characterized in that: Determining the target risk probability of each predicted rainfall level of the tested area in future rainfall events based on the waterlogging risk probability of each rainfall level includes: Determining, based on the waterlogging risk probability for each rainfall level, the initial risk probability for each predicted rainfall level in the tested area during future rainfall events; Obtaining a target risk probability of the predicted rainfall level according to the predicted probability of the predicted rainfall level and the initial risk probability of the predicted rainfall level; The predicted probability of the predicted rainfall level is the probability that the predicted rainfall amount of the future rainfall event will reach the predicted rainfall level.

5. The method for detecting waterlogging risk according to claim 1, wherein: Determining the waterlogging risk detection result of the test area in the future rainfall event based on the target risk probability of each predicted rainfall level, including: Superimposing the target risk probabilities of the predicted rainfall levels to obtain the probability of waterlogging in the test area in the future rainfall event; A waterlogging risk detection result of the tested area in the future rainfall event is determined according to the waterlogging probability.

6. The method for detecting waterlogging risk according to claim 1 or 5, characterized in that: Also includes: Determine that the waterlogging risk detection result indicates that the tested area has a waterlogging risk in the future rainfall event, and generate prompt information prompting the reconstruction of the drainage network in the tested area.

7. The method for detecting waterlogging risk according to claim 1, wherein: The rainfall amount is positively correlated with the rainfall level.

8. A waterlogging risk detection device, characterized in that: include: A rainfall event acquisition module is used to determine each historical rainfall event corresponding to any rainfall level based on the rainfall of each historical rainfall event in the test area; A first event determination module is configured to determine, from the historical rainfall events corresponding to the rainfall levels, risk rainfall events that may cause waterlogging in the test area; A second event determination module is configured to determine, from the risky rainfall events, target rainfall events whose accumulated water volume at the starting time is less than or equal to the predicted accumulated water volume at the starting time based on the predicted accumulated water volume of the future rainfall event at the starting time; a risk probability determination module, configured to determine the waterlogging risk probability of the rainfall level in the test area based on the target rainfall events corresponding to the rainfall level and the historical rainfall events corresponding to the rainfall level; a waterlogging risk detection module, configured to determine, based on the waterlogging risk probabilities for the respective rainfall levels, target risk probabilities for the respective predicted rainfall levels of future rainfall events in the tested area, and to determine, based on the target risk probabilities for the respective predicted rainfall levels, a waterlogging risk detection result for the tested area in the future rainfall events; The future rainfall event is the first rainfall event in the measured area in the future, and the predicted rainfall level is the predicted rainfall level of the future rainfall event.

9. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the processor executes the computer program, the waterlogging risk detection method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for detecting waterlogging risk according to any one of claims 1 to 7 is implemented.

Images