Urban inland inundation risk prediction method, medium and terminal

Abstract

The invention is suitable for the technical field of urban disaster prevention, and relates to an urban inland inundation risk prediction method, a medium and a terminal, and the method comprises the steps: S10, constructing a flood numerical simulation model, namely an SWMM model; s20, collecting rainfall data of a historical actual measurement session, importing an SWMM model, outputting a waterlogging node risk, obtaining a node risk sample, considering a rainfall characteristic value and disaster-causing continuity thereof, constructing a rainfall input mode, obtaining a rainfall sample, and constructing a training sample set of rainfall and node risks; s30, building a node risk prediction model based on a BP neural network by taking the session rainfall as input and the waterlogging node risk as output; and S40, quantifying the regional waterlogging risk by using an entropy weight method to obtain a regional waterlogging risk value so as to realize urban waterlogging risk prediction. The method is simple in process and convenient to operate, and the accuracy and timeliness of urban inland inundation risk prediction are effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban disaster prevention, and particularly relates to a method, medium and terminal for predicting urban waterlogging risk. Background Art

[0002] Urban waterlogging refers to the phenomenon that under the conditions of high-intensity and short-duration rainstorms, the runoff in the city exceeds the carrying capacity of the urban drainage system, resulting in the failure of rainwater to be discharged smoothly, thus causing large-scale waterlogging on the urban surface. Rapid and accurate prediction of urban waterlogging risk is a key link in the scientific development of waterlogging prevention work. Therefore, the development of effective urban waterlogging risk prediction methods is of great significance for improving the urban flood resistance ability, ensuring public safety and reducing economic losses.

[0003] The current urban waterlogging risk prediction model has the following two limitations: First, it does not fully consider the continuity of the rainfall disaster-causing process and the comprehensiveness of input parameters. For example, rainfall intensity, total rainfall, peak coefficient, etc. will have an important impact on the degree and duration of waterlogging, resulting in low accuracy of urban waterlogging prediction; Second, the operation time of traditional numerical models is relatively long, which cannot meet the requirements of urban waterlogging risk prediction in terms of timeliness. The patent with the publication number CN106373070B provides a four-prediction method for coping with urban rainstorm waterlogging. Step 1: Use urban rainstorm waterlogging estimation to perform statistical downscaling on different global climate model (GCM) data in the area where the city is located to obtain the estimated daily rainfall; Determine whether extreme precipitation occurs based on the estimated daily rainfall: If extreme precipitation occurs, go to Step 2 for urban rainstorm waterlogging prediction, otherwise, continue with urban rainstorm waterlogging estimation; Step 2: Further predict whether the city will experience rainstorm waterlogging based on the estimated daily rainfall obtained in Step 1: If it is predicted that rainstorm waterlogging will occur, go to Step 3 for urban rainstorm waterlogging warning; Otherwise, return to Step 1 to continue with urban rainstorm waterlogging estimation; Step 3: Process according to the prediction result in Step 2: Integrate and fuse the urban rainstorm waterlogging data resources, and use the established "entity-relationship" deduction model to deduce the rainstorm waterlogging information in key urban areas. The key urban areas include the banks of urban rivers, low-lying areas, key industrial parks, and waterlogging-prone points, to obtain urban rainstorm waterlogging warning information, and then enter Step 4 for further processing of the urban rainstorm waterlogging plan; Otherwise, return to Step 2 to continue with urban rainstorm waterlogging prediction; Step 4: Make a judgment based on the urban rainstorm waterlogging warning information obtained in Step 3: If the urban rainstorm waterlogging warning information reaches the level that requires response, digitize the traditional paper and graphic plans, establish a scenario plan, and conduct qualitative and quantitative analysis on the scenario plan based on a visualization platform to finally form an urban rainstorm waterlogging response plan; Otherwise, return to Step 3 to continue with rainstorm waterlogging warning. Although this patent realizes the prediction of rainstorm waterlogging, it only makes predictions based on daily rainfall, with low accuracy, and has the same drawbacks as the existing technology.

[0004] Therefore, how to provide an urban waterlogging risk prediction method that can take into account both prediction accuracy and timeliness is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide an urban waterlogging risk prediction method to solve the problems of low accuracy and timeliness of the urban waterlogging risk prediction model in the prior art; In addition, the present invention also provides an urban waterlogging risk prediction medium and a terminal.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for predicting urban waterlogging risk, comprising the following steps:

[0008] S10, constructing a flood numerical simulation model, namely, the SWMM model;

[0009] S20, collecting historical measured rainfall data, importing it into the SWMM model and outputting the waterlogging node risk, obtaining node risk samples, considering rainfall characteristic values ​​and its disaster-causing continuity, constructing a rainfall input model, obtaining rainfall samples, and constructing a training sample set of rainfall and node risk;

[0010] S30, taking rainfall events as input and waterlogging node risk as output, a node risk prediction model is constructed based on BP neural network;

[0011] S40. Quantify the regional waterlogging risk by using the entropy weight method to obtain the regional waterlogging risk value, so as to realize the urban waterlogging risk prediction.

[0012] Furthermore, the specific steps of step S10 are as follows:

[0013] S101. Based on the urban drainage network data, ArcGIS software is used to extract the length, diameter, buried depth of the starting and ending points of the pipelines, the ground elevation of the inspection wells, and the starting and ending point numbers of the pipelines. Based on the spatial topological relationship of the drainage network, necessary nodes are added, branches and secondary pipelines are deleted, and the urban drainage network is generalized;

[0014] S102, using the analysis tool of ArcGIS software to draw Thiessen polygons according to the node distribution, and finally clipping the details according to actual needs to obtain the sub-catchment area, and obtaining the imperviousness and slope of the catchment area by performing GIS intersection operation on the sub-catchment area layer, urban land use type map, and DEM elevation map;

[0015] S103, using the tools provided by the SWMM model to establish a link to interact with the Excel table data, entering the data of the pipeline, node, and catchment area into the corresponding area of ​​the Excel table, saving the Excel table to complete the establishment of the INP file, and completing the modeling after supplementing some data in the SWMM model;

[0016] S104, inputting the measured rainfall into the SWMM model, and verifying the rationality of the model by comparing the outlet flow process line simulated by the SWMM model with the measured flow process line.

[0017] Furthermore, the specific steps of step S20 are as follows:

[0018] S201. Collect urban rainfall data for each event, identify and screen rainfall causing disasters, import the selected heavy rainfalls into the SWMM model to obtain the overflow volume of regional nodes, and calculate the risk of regional nodes using the following formula:

[0019]

[0020] where V Ti is the total overflow water volume of the i-th node in the region, and the total overflow water volume of each node is obtained from the simulation results of the SWMM model; n i is the number of stormwater wells around the sub-catchment draining into the i-th node, and its value is the ratio of the perimeter of the sub-catchment to the setting distance of the stormwater wells; ω i is the weight coefficient of the i-th node, and its value is the percentage of the area of the overlapping part of the Thiessen polygon of the node and the region in the total area of the region.

[0021] S202. Determine the rainfall input mode, which includes cumulative rainfall + rainfall characteristics, cumulative rainfall, hourly rainfall + rainfall characteristics, hourly rainfall, and rainfall characteristics, to form a rainfall sample training set for rainfall and node risk. The rainfall characteristics include total rainfall, average rainfall intensity, peak multiple ratio, peak rainfall amount, and rainfall peak coefficient.

[0022] Furthermore, the specific steps of step S30 are as follows:

[0023] S301. Build and train a BP neural network model. Relying on the MATLAB software, build a BP neural network by calling the newff function, perform normalization processing on the rainfall and node risk training sample set, select the training function and activation function, and determine the number of iterations, learning rate, number of hidden layers, and number of neurons in the hidden layer.

[0024] S302. By comparing the performance of the model under each rainfall input mode, select the best input mode as the input of the BP prediction model.

[0025] Furthermore, the specific steps of step S40 are as follows:

[0026] S401. Use the natural disaster risk expression to evaluate the risk of urban waterlogging. Select node risk, ground slope, elevation, and impervious rate as hazard indicators, and population, GDP per unit area, and road network as vulnerability indicators.

[0027] S402. Perform normalization processing on each index value, determine the weight of each evaluation index by the entropy weight method, and calculate the weighted sum to obtain the urban waterlogging risk value of the region.

[0028] In a second aspect, the present invention further provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the method as described above.

[0029] In a third aspect, the present invention further provides an electronic terminal, comprising: a processor and a memory; the memory is used for storing a computer program, and the processor is used for executing the computer program stored in the memory so that the terminal executes the method as described above.

[0030] Compared with the prior art, the urban waterlogging risk prediction method, medium and terminal provided by the present invention have at least the following beneficial effects:

[0031] The current urban waterlogging risk prediction model has the following two limitations: First, it does not fully consider the continuity of the rainfall disaster-causing process and the comprehensiveness of input parameters. For example, rainfall intensity, total rainfall, peak coefficient, etc. will all have an important impact on the degree and duration of waterlogging, resulting in low accuracy of urban waterlogging prediction; Second, the running time of traditional numerical models is relatively long, which cannot meet the requirements of the timeliness of urban waterlogging risk prediction. The process of the present invention is simple and easy to operate. By considering the continuity of the rainfall disaster-causing process, a new rainfall input mode is constructed, and a mapping relationship between rainfall and waterlogging is established based on the neural network model, so as to construct a waterlogging risk prediction model that takes into account both accuracy and timeliness, effectively improving the accuracy and timeliness of urban waterlogging risk prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the solution of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0033] Figure 1 It is a flowchart of a method for predicting urban waterlogging risk provided by an embodiment of the present invention;

[0034] Figure 2 It is a schematic diagram of the land use type of a certain city in a method for predicting urban waterlogging risk provided by an embodiment of the present invention;

[0035] Figure 3 It is a schematic diagram of the drainage pipe network and sub-catchments of a certain city in a method for predicting urban waterlogging risk provided by an embodiment of the present invention;

[0036] Figure 4 It is a comparison diagram of the measured and simulated drainage outlet flow during a measured rainfall in a certain city in a method for predicting urban waterlogging risk provided by an embodiment of the present invention;

[0037] Figure 5Comparison chart of measured and simulated drainage outlet flow in another measured rainfall in a certain city in the urban waterlogging risk prediction method provided by the embodiment of the present invention;

[0038] Figure 6 Schematic diagram of flood-prone areas in a certain city in the urban waterlogging risk prediction method provided by the embodiment of the present invention;

[0039] Figure 7 Risk prediction result chart of the first node in the flood-prone area with the highest flood risk in a certain city in the urban waterlogging risk prediction method provided by the embodiment of the present invention;

[0040] Figure 8 Risk prediction result chart of the second node in the flood-prone area with the highest flood risk in a certain city in the urban waterlogging risk prediction method provided by the embodiment of the present invention;

[0041] Figure 9 Risk prediction result chart of the third node in the flood-prone area with the highest flood risk in a certain city in the urban waterlogging risk prediction method provided by the embodiment of the present invention;

[0042] Figure 10 Risk prediction result chart of the fourth node in the flood-prone area with the highest flood risk in a certain city in the urban waterlogging risk prediction method provided by the embodiment of the present invention. Detailed implementation manners

[0043] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. The preferred embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the understanding of the disclosure of the present invention more thorough and comprehensive.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0045] The present invention provides an urban waterlogging risk prediction method, which is applied to the urban waterlogging risk prediction process considering the continuity of the rainfall disaster-causing process. The urban waterlogging risk prediction method includes the following steps:

[0046] S10. Construct a flood numerical simulation model, namely the SWMM model; S20. Collect historical measured rainfall data, import it into the SWMM model and output the waterlogging node risk to obtain the node risk sample, consider the rainfall characteristic value and its disaster continuity, construct the rainfall input pattern, obtain the rainfall sample, and build a training sample set of rainfall and node risk; S30. Take the rainfall as input and the waterlogging node risk as output, and build a node risk prediction model based on the BP neural network; S40. Quantify the regional waterlogging risk by the entropy weight method, and obtain the regional waterlogging risk value to realize the urban waterlogging risk prediction.

[0047] The method of the present invention has a simple process and is easy to operate, and effectively improves the accuracy and timeliness of urban waterlogging risk prediction.

[0048] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings.

[0049] The present invention provides a method for predicting urban waterlogging risk, which is applied to the process of predicting urban waterlogging risk by considering the continuity of the rainfall disaster process. Figures 1 to 10 In this embodiment, the method for predicting urban waterlogging risk in a certain city includes the following steps:

[0050] S10. Construct a SWMM model for numerical simulation of floods in a certain city.

[0051] Specifically, in this embodiment, the specific steps of constructing a city SWMM model are as follows:

[0052] S101. Based on the spatial topological relationship of the drainage network, necessary nodes are added, branches and secondary pipelines are deleted, and the urban drainage network is generalized, such as Figure 3 As shown in the figure, the attribute information of pipelines and nodes was extracted using ArcGIS software based on the drainage network data of a certain city.

[0053] S102, using ArcGIS analysis tools to draw Thiessen polygons based on node distribution, and finally clipping the details according to actual needs to obtain sub-catchment areas, such as Figure 3 As shown in the figure, the watershed layer is intersected with the urban land use status map and the DEM elevation map by GIS, and the attribute information such as the watershed impervious rate and slope are obtained.

[0054] S103. Use the tools provided by SWMM to establish a link for data interaction in the Excel table, enter the data of pipelines, nodes, catchment areas, etc. into the corresponding areas in the Excel table, save the Excel table to complete the creation of the INP file, and complete the modeling after supplementing some data in SWMM.

[0055] S104. Input the measured rainfall of two cases in a certain city into the SWMM model of the city to obtain the corresponding drainage outlet flow process. Compare the flow process line of Drainage Outlet 1 with the measured data. As Figure 4 shown in Figure 5 , the two are basically in agreement, indicating good applicability of the model.

[0056] S105. Input rainfall with different return periods into the SWMM model of a certain city. By calculating the node risks, 8 flood-prone areas are extracted, as Figure 6 shown.

[0057] S20. Construct a training sample set of rainfall (input) - node risk (output) for a certain city.

[0058] Specifically, in this embodiment, the specific steps of step S20 are as follows:

[0059] S201. Collect the precipitation data of a certain market, identify the disaster-causing rainfall, and find that the main type of disaster-causing rainfall in this city is short-duration heavy rain within three hours, and then conduct screening. Import the screened heavy rain into the SWMM model to obtain the overflow water volume of each node in the flood-prone areas, and calculate the node risks of each area using the following formula.

[0060]

[0061] where V Ti is the total overflow water volume of the i-th node in the area, and the total overflow water volume of each node is obtained from the simulation results of the SWMM model; n i is the number of rainwater wells around the sub-catchment area draining into the i-th node, and its value is the ratio of the perimeter of the sub-catchment area to the set distance of the rainwater wells; ω i is the weight coefficient of the i-th node, and its value is the percentage of the area of the overlapping part of the Thiessen polygon of the node and the area of the region in the total area of the region.

[0062] S202. Determine the rainfall input mode. The original format of the rainfall sample set is generally the rainfall process in minute-by-minute time periods. Considering the continuity of the rainfall disaster-causing process and the main characteristics of the rainfall pattern (such as total rainfall, average rainfall intensity, peak multiple ratio, peak rainfall, rainfall peak coefficient, etc.), five rainfall input modes are finally formed, as shown in Table 1. Each of the five input mode rainfall samples derived from each original rainfall sample corresponds to the same node risk sample, thus constituting the rainfall - node risk training sample set.

[0063] Table 1

[0064]

[0065] S30. Build a node risk prediction model for a certain city based on the BP neural network.

[0066] Specifically, in this embodiment, the specific construction steps of step S30 are as follows:

[0067] S301. Build and train a BP neural network model. Relying on the MATLAB software, a BP neural network is built by calling the newff function, the data (rainfall-node risk training sample set) is normalized, training functions and activation functions are selected, and parameters such as the number of iterations, learning rate, number of hidden layers, and number of neurons in the hidden layer are determined.

[0068] S302. By comparing the performance of the model under various rainfall input patterns, select the best input pattern as the input of the BP prediction model.

[0069] S303. Evaluate the BP model with four nodes in the most waterlogging-prone area 7 in a certain city. The prediction results are as Figures 7 to 10 shown. The predicted values and the actual values fit well, and the model has high reliability.

[0070] S40: Quantify the waterlogging risk in a certain city's waterlogging-prone area.

[0071] Specifically, in this embodiment, in step S4, the entropy weight method is used to quantify the risk of the waterlogging-prone area in a certain city. The specific steps are as follows:

[0072] S401. Use the natural disaster risk expression (risk = hazard + vulnerability) proposed by Marskey in 1989 to evaluate the risk of urban waterlogging. For urban waterlogging, disasters are the result of the combined action of society and nature. Therefore, node risk, ground slope, elevation, and impervious rate are selected as hazard indicators, and population, GDP per unit area, and road network are selected as vulnerability indicators.

[0073] S402. Normalize each index value, determine the weights of each evaluation index through the entropy weight method, and perform weighted summation calculation to obtain the regional waterlogging risk value.

[0074] S403. Input the rainfall with return periods of 1 year, 3 years, 5 years, 10 years, and 20 years in a certain city into the urban node risk prediction model of this invention's embodiment to obtain the node risk prediction results of each waterlogging-prone area in the city. Normalize each index value of the waterlogging-prone area, determine the weights of each evaluation index through the entropy weight method, and perform weighted summation calculation to obtain the waterlogging risk values of each waterlogging-prone area, as shown in Table 2.

[0075] Table 2

[0076]

[0077] Thus, the risk prediction of urban waterlogging based on the continuity of the rainfall disaster-causing process for different return periods is completed.

[0078] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, any one of the methods in this embodiment is implemented.

[0079] An embodiment of the present invention also provides an electronic terminal, including: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the terminal executes any one of the methods in this embodiment.

[0080] For the computer-readable storage medium in this embodiment, those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to the computer program. The foregoing computer program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: various media such as ROM, RAM, magnetic disk or optical disc that can store program codes.

[0081] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store a computer program, the communication interface is used for communication, and the processor and the transceiver are used to run the computer program so that the electronic terminal executes each step of the above method.

[0082] Compared with the prior art, the current urban waterlogging risk prediction model has the following two limitations for the urban waterlogging risk prediction method, medium and terminal described in the above embodiments: First, it does not fully consider the continuity of the rainfall disaster-causing process and the comprehensiveness of input parameters. For example, rainfall intensity, total rainfall, peak coefficient, etc. will all have an important impact on the degree and duration of waterlogging, resulting in low accuracy of urban waterlogging prediction; Second, the running time of traditional numerical models is relatively long, which cannot meet the requirements of urban waterlogging risk prediction in terms of timeliness. The process of the present invention is simple and easy to operate. By considering the continuity of the rainfall disaster-causing process, a new rainfall input mode is constructed, and a mapping relationship between rainfall and waterlogging is established based on the neural network model, so as to construct a waterlogging risk prediction model that takes into account both accuracy and timeliness, effectively improving the accuracy and timeliness of urban waterlogging risk prediction.

[0083] Obviously, the embodiments described above are only the preferred embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosed content of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing specific embodiments, or perform equivalent replacements on some of the technical features. Any equivalent structure made by using the content of the specification and drawings of the present invention, directly or indirectly applied in other related technical fields, is similarly within the scope of the patent protection of the present invention.

Claims

1. A method for predicting urban waterlogging risk, characterized in that: The following steps are involved: S10, constructing a flood numerical simulation model, namely, the SWMM model; S20, collecting historical measured rainfall data, importing it into the SWMM model and outputting the waterlogging node risk, obtaining node risk samples, considering rainfall characteristic values ​​and its disaster-causing continuity, constructing a rainfall input model, obtaining rainfall samples, and constructing a training sample set of rainfall and node risk; S30, taking rainfall events as input and waterlogging node risk as output, a node risk prediction model is constructed based on BP neural network; S40. Quantify the regional waterlogging risk by using the entropy weight method to obtain the regional waterlogging risk value, so as to realize the urban waterlogging risk prediction.

2. The urban waterlogging risk prediction method according to claim 1, characterized in that: The specific steps of step S10 are as follows: S101. Based on the urban drainage network data, ArcGIS software is used to extract the length, diameter, buried depth of the starting and ending points of the pipelines, the ground elevation of the inspection wells, and the starting and ending point numbers of the pipelines. Based on the spatial topological relationship of the drainage network, necessary nodes are added, branches and secondary pipelines are deleted, and the urban drainage network is generalized; S102, using the analysis tool of ArcGIS software to draw Thiessen polygons according to the node distribution, and finally clipping the details according to actual needs to obtain the sub-catchment area, and obtaining the imperviousness and slope of the catchment area by performing GIS intersection operation on the sub-catchment area layer, urban land use type map, and DEM elevation map; S103, using the tools provided by the SWMM model to establish a link to interact with the Excel table data, entering the data of the pipeline, node, and catchment area into the corresponding area of ​​the Excel table, saving the Excel table to complete the establishment of the INP file, and completing the modeling after supplementing some data in the SWMM model; S104, inputting the measured rainfall into the SWMM model, and verifying the rationality of the model by comparing the outlet flow process line simulated by the SWMM model with the measured flow process line.

3. The urban waterlogging risk prediction method according to claim 1, characterized in that: The specific steps of step S20 are as follows: S201, collect urban market precipitation data, identify and screen disaster-causing rainfall, import the screened rainstorms into the SWMM model, obtain the overflow of regional nodes, and calculate the regional node risk using the following formula: Among them, V Ti is the total overflow water volume of the ith node in the region, and the total overflow water volume of each node is obtained by the simulation results of the SWMM model; n i is the number of rainwater wells around the subcatchment that drains into the i-th node, and its value is the ratio of the perimeter of the subcatchment to the distance between the rainwater wells; ω i is the weight coefficient of the ith node, and its value is the percentage of the area of ​​the overlapping part of the node Thiessen polygon and the region to the total area of ​​the region; S202. Determine a rainfall input mode, wherein the rainfall input mode includes cumulative rainfall + rainfall characteristics, cumulative rainfall, time period rainfall + rainfall characteristics, time period rainfall and rainfall characteristics, constituting a rainfall sample training set for rainfall and node risks. The rainfall characteristics include total rainfall, average rainfall intensity, peak-to-peak ratio, peak rainfall, and rain peak coefficient.

4. The urban waterlogging risk prediction method according to claim 1, characterized in that: The specific steps of step S30 are as follows: S301, build and train the BP neural network model, rely on MATLAB software, build the BP neural network by calling the newff function, normalize the rainfall and node risk training sample sets, select the training function and activation function, and determine the number of iterations, learning rate, number of hidden layers, and number of neurons in the hidden layer; S302: By comparing the performance of the models under various rainfall input modes, the best input mode is selected as the input of the BP prediction model.

5. The urban waterlogging risk prediction method according to claim 1, characterized in that: The specific steps of step S40 are as follows: S401. Use natural disaster risk expressions to assess urban waterlogging risk, select node risk, ground slope, elevation, and imperviousness as hazard indicators, and population, GDP per unit area, and road network as vulnerability indicators; S402, normalize the index values, determine the weights of the evaluation indexes by using the entropy weight method, and perform weighted sum calculation to obtain the regional waterlogging risk value.

6. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

7. An electronic terminal, characterized in that: include: Processor and memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes the method according to any one of claims 1 to 5.

Images